Computer Vision and Pattern Recognition 130
☆ Back to Pixels: Lifting Latent Video Diffusion into Pixel Space
Yanke Song, Shuo Chen, Chris Rockwell, Guande He, Nan Liu, Desai Xie, Tianpei Gu, Yong-Hyun Park, Aryan V S, Dennis Yuren Menn, Ming-Hsuan Yang, Xun Huang
Frontier video generation models work in a VAE latent space that compresses video in both space and time. This compression made training tractable, but now limits high-resolution, controllable, and interactive generation -- doubling resolution quadruples token count, conditions apply to groups of frames rather than individual ones, and each response must wait for the rest of its multi-frame block. We present PixVid, a simple yet effective method that lifts a pretrained latent video model to directly generate every frame in pixel space. PixVid keeps the diffusion transformer and replaces the VAE decoder with a convolutional head that renders its outputs into pixels. We find that predicting clean video rather than velocity, together with a larger noise shift, is key to quality. Finetuned from Wan2.1-14B, PixVid is competitive with the latent baseline within $\sim 1.3$M training clips. Working framewise in pixels, it generates at $2\times$ higher resolution with the same per-frame token count, keeps depth-conditioned video editing aligned under fast motion, and responds to each action at the next frame. Project page: https://chenshuo20.github.io/PixVid/
comment: Project Page: https://chenshuo20.github.io/PixVid/
☆ Unbundling Multimodal Vision Encoding
Ellis Brown, Chris Hoang, Shengyi Qian, Shang-Wen Li, Mike Lewis, Marjan Ghazvininejad, Luke Zettlemoyer, Lucas Beyer, Mengye Ren, Jakob Verbeek, Saining Xie, Hu Xu
How multimodal language models should encode visual input remains up for debate. Most models use a pretrained vision encoder, while some recent recipes train the encoder from scratch with the language model or remove it entirely. Vision encoders bundle together three design choices: a visual prior from upstream pretraining, dedicated vision capacity, and pre-fusion placement of that capacity (i.e., before the multimodal backbone). Because published models also differ in data, architecture, and compute, comparing them cannot isolate the effect of each choice. We disentangle these choices via a controlled framework where we pretrain all models from scratch on common data and compare them on visual question answering and language benchmarks at matched lifetime compute. We find that the visual prior is not necessary; an encoder trained jointly from scratch is on par with a pretrained one, without a separate training stage. What matters instead is dedicated vision capacity: an encoder-free model trails on visual understanding even when its backbone is enlarged by the encoder's parameter count. Yet that capacity need not precede multimodal fusion, as vision-only modules added in-fusion to early backbone layers perform comparably to a pre-fusion encoder and can lower time to first token. These results also hold across backbone sizes, when encoder pretraining is treated as free, and when starting from a pretrained language model. Overall, our findings favor dedicated vision capacity trained jointly with the backbone and placed early within it.
comment: Project page: https://unbundling.github.io
☆ TTB: Test-time Baking for Efficient Rendering of Decoder-only View Synthesis Models NeurIPS 2026
Chin-Yang Lin, Ryo Hachiuma, Jaesung Choe, Min-Hung Chen, Yu-Chiang Frank Wang, Yen-Yu Lin, Wei-Chen Chiu, Yu-Lun Liu, Cheng Sun
Decoder-only Large View Synthesis Models (LVSMs) with KV-cache have recently achieved state-of-the-art quality by regressing novel views from neural networks without reconstructing 3D geometry, while Test-Time Training (TTT) layers save computation but sacrifice quality. We propose Test-Time Baking (TTB), a strategy that learns to bake global scene information from cross-view attention into a lightweight MLP, enabling fast and high-fidelity novel-view decoding. The first key is to repurpose the fast-weight MLP of TTT to learn the cross-view attention mapping explicitly from its input-output pairs, rather than replacing attention entirely, achieving the same rendering efficiency as TTT while delivering substantially higher quality. The second key addresses a modality misalignment between context prefilling and novel-view rendering: we disentangle context tokens into camera-only context rendering tokens to serve as TTB input, and context ground-truth tokens with image content to serve as the TTB mapping target, ensuring the baked MLPs are relieved of the modality generalization challenge between context and novel-view processing. Our method achieves up to **23x** rendering speedup with quality on par with the state-of-the-art across several benchmarks. Project page: https://test-time-baking.github.io
comment: NeurIPS 2026 (Spotlight). Project page: https://test-time-baking.github.io
☆ ProtoVis-Nav: Prototypical Visual Imagination for Visual-Spatial Aligned UAV Navigation
Vision language navigation (VLN) is very challenging for Unmanned Aerial Vehicles (UAVs) since small errors in 3D motion prediction can lead to navigation errors and collisions. Unable to fathom execution outcomes in 3D environments, existing methods often struggle to predict accurate navigation waypoints from language instructions. To solve this, we introduce the Prototypical Visual Imagination framework, shortened as ProtoVis-Nav, which forces the model to imagine an outcome-associated local visual representation for the intended next UAV movement. Specifically, we predict feature prototypes and use their linear combination to estimate the visual feature of the target region. In this manner, the VLN model first imagines the local visual representation of the next intended location, and then uses predicted prototypes as context for generating the final action, promoting alignment between visual context and waypoint outputs. Extensive experiments demonstrate the effectiveness of the ProtoVis-Nav framework and show substantial improvements over existing baselines.
☆ K2V: Generative Keyframing For Physically Plausible Video Generation
Video diffusion models are able to generate samples with high-quality appearance and textural qualities. However, content produced by these methods is known to exhibit significant physical errors. This is because, in a classical video generation setting, all physical reasoning must take place simultaneously and exclusively within the diffusion process.We propose to resolve these issues by redistributing the task of physical reasoning onto a language model-backed keyframe generation module, which allows for explicit reasoning in language, and fine-grained control over the appearance of the video at important moments in the video. We then condition a video generation model with these keyframes, reducing the role of the diffusion model to the simpler task of interpolating between physically-correct intermediate states, rather than inferring an entire sequence with diffusion. Results demonstrate that our approach significantly outperforms existing approaches, achieving state-of-the-art performance on the Physics-IQ Verified benchmark and PhyGenBench.
☆ BrickFlow: Connectivity-Guided Brick Reconstruction NeurIPS 2026
Given a set of LEGO parts and an image of an assembled object, our task is to infer the pose of each part to assemble the object. In contrast with traditional 3D reconstruction, doing so requires satisfying discrete physical constraints, which provide structure but render the problem combinatorial and highly sensitive to small pose errors. To learn a prior over how parts fit together, we train a set-conditioned SE(3) flow model on procedurally generated assemblies. We then finetune this model to inject image conditioning, using 1% as much data. However, while the learned flow captures global structure, it does not enforce connectivity. To address this, we introduce an analytic connectivity-field guidance to pull predictions toward connected assemblies at test time. Through this, we demonstrate improved build quality with model size, pretraining, and guidance. We make our models and data available for research purposes at https://kulits.github.io/BrickFlow/
comment: NeurIPS 2026
☆ Spatial Foundry: A Multimodal Dataset for Spatial Intelligence
Spatial intelligence requires connecting visual observations to the geometry, objects, and persistent state of an environment. Images alone leave these relationships implicit, while disconnected annotations make it difficult to study appearance, camera motion, and scene changes under a common reference. We introduce Spatial Foundry, a collection of 5,280 furnished residential 3D scene configurations with editable assets and aligned visual, geometric, and temporal data. Its modalities include floor plans, perspective and panoramic RGB, metric depth, point clouds, meshes, instance-level 3D bounding boxes, and camera trajectories paired with video. Shared scene coordinates link observations across views and time; editable source scenes enable additional camera paths and controlled object changes. We describe the data representation and construction workflow, and discuss applications in floor-plan-conditioned 3D generation, geometric reconstruction, instance-level perception, camera-conditioned video generation, shared world modeling, and scene editing. Controlled camera and scene variations support spatial perception and world-model research. The public project page provides videos and an interactive colored point-cloud preview.
comment: Project Page: https://ranolabs.com/Blog/spatial-foundry
☆ FLEX-LBP: Learning Local Binary Patterns with Exact Binary Inference for Texture Classification
Local Binary Patterns (LBP) remain among the most widely used texture descriptors, from industrial inspection to medical imaging. Yet their radius, sampling angles, color transform and thresholds are fixed by hand for each dataset. We present FLEX-LBP, which learns these parameters by backpropagation while every forward pass computes exact binary codes and hard histogram counts. A straight-through estimator acts only in the backward pass, so the deployed descriptor is an interpretable hard-binary LBP with 90 learnable parameters and unchanged inference cost. In a controlled ablation on six color texture benchmarks, the component worth learning is dataset-dependent. One configuration, applied unchanged everywhere, learns the geometry and thresholds with the color transform frozen. It improves the nearest-neighbor accuracy of the identical fixed descriptor by 3.55 points on average over five seeds, significantly on five of six benchmarks and never significantly worse. With a linear discriminant readout, it gains 0.9 points; the fully learned descriptor gains 2.7 and has the highest mean. FLEX-LBP is more accurate than five classical LBP-family descriptors, PCANet and a neural LBP on five of six benchmarks. Its mean nearest-neighbor accuracy exceeds that of frozen ImageNet ResNet-50 features (23.5 million parameters). On the Kather colorectal cancer histology benchmark, the fully learned descriptor reaches 93.8%, significantly above the best classical descriptor (Completed LBP, 92.8%). The learned radii adapt in directions that recur across seeds. Noise drowns single-pixel comparisons and illuminant change flips inter-channel ones; median sampling or gray-world normalization restores 96-97%. The code is available as a drop-in PyTorch layer at https://github.com/medalimoussa/FLEX-LBP.
comment: 21 pages, 5 figures, 8 tables. Code: https://github.com/medalimoussa/FLEX-LBP
☆ Consistent Video-to-Video Translation via Explicit Correspondences ECCV 2026
Gaurav Parmar, Zhengqi Li, Richard Zhang, Jun-Yan Zhu, Srinivasa Narasimhan, Eli Shechtman, Yotam Nitzan
Interactive video-to-video applications require real-time generation while maintaining long-range temporal consistency. However, recent methods achieve speed by restricting attention to a fixed-length causal window. This limits long-range consistency whenever relevant history falls outside the context window. To address this limitation, we introduce vid2vid-long, a correspondence-based approach grounded in a simple observation. Video-to-video translation provides an explicit cross-frame reference: the regions that match in the input video should match in the output video too. Concretely, we compute patch-level correspondences in the input, guiding retrieval of previously generated outputs in the corresponding regions, and condition the model on these patches through correspondence cross-attention. Our experiments demonstrate that vid2vid-long consistently improves long-range temporal consistency across recent video models while preserving generation quality and real-time performance.
comment: ECCV 2026. Project Website: https://www.cs.cmu.edu/~vid2vid-long/
☆ Objects as Addresses: Instance Binding in Vision Transformer Attention
Humans naturally parse visual scenes into coherent individual objects rather than scattered features, a capability cognitive science calls object binding. While recent work shows that pretrained Vision Transformers (ViTs) intrinsically acquire this ability, how native self-attention computations mechanistically keep object instances separate remains unresolved. By studying this across eight pretrained ViTs, we found that instance separation is expressed more strongly in attention routing than in token content: selected query-key relations distinguish individual objects more reliably than value similarity or residual representations. Inside heads, patches belonging to one individual tend to read from a common image region, which we term their attention address. Visually distinct parts of one person share an address, whereas identical duplicated instances receive separate, movable addresses. These shared destinations arise in part because query interactions with the spatial structure of keys direct attention toward consistent regions for one object and elsewhere for another. In DINOv2 and most other backbones we tested, editing this query component at the selected head redirects its attention address toward another individual. Grouping patches by this query component separates same-category instances, reaching .615 FG-ARI versus .289 for the same clustering of final-layer features, which tend to merge them.
☆ MALNISI-Infused Graph Construction and Weighted Contrastive Loss based GAT for Unsupervised Representation Learning in HSI
Hyperspectral images (HSI) possess discriminating spectral characteristics with respect to land utilization due to high-dimensional information. Recently, unsupervised representation learning using graph neural networks has unveiled these utilization-specific complex spectral relations. This article has three main contributions. First, it proposes an index, the Mean Absolute Local Neighborhood Inverse Spectral Index (MALNISI), which measures the similarity between two pixels based on their neighborhood information. The index is used to construct a weighted adjacency matrix, giving the HSI a graph structure. Second, it proposes an upgraded Graph Attention Network that learns compact latent representations of hyperspectral data from the constructed graph. In this network, an enhanced multi-head attention mechanism incorporates the MALNISI weights as an auxiliary feature to emphasize learned attention between neighboring pixels. To alleviate over-smoothing and preserve fine-grained spectral details, the architecture integrates residual connections, skip projections, and feature normalization, which together enhance representation capability, stabilize learning, and accelerate convergence. Third, it proposes a weighted contrastive loss function that incorporates neighborhood-aware MALNISI between connected pixels, encouraging strongly connected pixels in the high-dimensional space to remain close in the latent space, and weakly connected pixels to stay apart. The resulting discriminative low-dimensional embeddings are then clustered using spectral clustering. Experimental analysis on six benchmark HSI datasets demonstrates the efficiency and necessity of the three aforementioned contributions, and the comparative study highlights superior performance over fourteen state-of-the-art approaches.
☆ eDMD: Endpoint Distribution Matching for Few-Step Diffusion Distillation
Distribution matching distillation (DMD) has emerged as a leading paradigm for distilling diffusion models into few-step students, where an auxiliary fake score model tracks the student's distribution and the discrepancy between the teacher and fake predictions drives the student toward the teacher. In multi-step distillation, although only the student's final output is delivered, existing methods train the fake model on intermediate states or across all denoising steps. Consequently, the student's update signal becomes entangled with intermediate distributions, which can distort the matching direction and remain nonzero even after the student's final outputs already match the teacher distribution. To resolve this mismatch, we propose Endpoint Distribution Matching Distillation (eDMD), which trains a single shared fake model exclusively on the student's final outputs and uses it to correct every denoising step. Our theoretical analysis proves that this endpoint reference guarantees that the distribution matching term vanishes if and only if the student's final output distribution matches the teacher's, regardless of intermediate distributions. Beyond the fake reference, we further strengthen teacher supervision through a precise teacher (PT) that replaces single-step guided predictions with short multi-step trajectories on the final noise interval, paired with a calibrated query density for intermediate steps. Experiments on SD3-Medium, FLUX.2-klein (4B and 9B), and Qwen-Image-2.1 demonstrate that eDMD consistently outperforms existing distribution matching methods for 4-step generation.
comment: Code Link: see https://github.com/Westlake-AGI-Lab/eDMD
☆ Consensus Capture in Reference-Based UAV Video Stabilization: A Benchmark on Dynamic Urban Scenes
Reference-frame stabilization registers every frame of a hovering-drone video onto one reference frame, keeping object positions comparable. The registration pipelines it relies on are selected on feature-matching benchmarks in which only the camera moves. Over a busy intersection, vehicles moving together supply mutually consistent correspondences that can draw the robust homography estimate toward their motion: consensus capture, which inlier statistics cannot flag. A two-part benchmark measures it and the object masking meant to prevent it. A static protocol scores ten pipelines under exactly known homographies on 4K aerial frames, tuned on a site-disjoint split at a matched resolution and feature budget. A dynamic protocol scores them at an offline and a real-time budget on hovers selected from the video alone to span the amount and coherence of moving foreground, plus a stationary control. It measures background alignment, object position error and the share of a moving group's displacement an estimate follows, against a pseudo-ground-truth reference audited with manually annotated ground control points, which also score every candidate. On static scenes, tuned AKAZE, RootSIFT, SIFT and BRISK lead, retuning the matcher thresholds raises the mean average accuracy (mAA) by at most 0.0062, and masking costs less than 0.005. Unmasked ORB is captured on 28 of 304 real-time dynamic pairs and on 6 offline; its real-time captured estimates have a median corner error of 18.8 px yet typical inlier counts and ratios. Masking removes capture for every pipeline and leaves the leading pipelines unchanged. Code, data and results are released.
comment: 50 pages, 17 figures, 24 tables. Code at https://github.com/rfonod/stabilo-benchmark and data at https://doi.org/10.5281/zenodo.23623540 and https://doi.org/10.5281/zenodo.15050577
☆ Neural Render Proxies for Interactive and Differentiable Lighting
Sergio Sancho, Alexander Rath, Marco Manzi, Pascal Chang, Amit H. Bermano, Derek Nowrouzezahrai, Markus Gross, Marios Papas
Within the CG animation production pipeline, the challenges artists tackle in Lighting can be immense. Even minor adjustments require re-rendering massive scenes with slow offline renderers; global illumination has to be sampled and complex shaders have to be evaluated, leading to iteration times of minutes to hours per frame. To accelerate this process, we introduce a novel neural render proxy (NRP) that enables differentiable relighting of static scenes with fixed camera and materials at interactive rates. Our main insight is the decoupling of traditional rendering into path sampling and emission computation. From a single, light-agnostic render pass, we collect light transport data in the form of path samples. This enables rapidly sampling lighting conditions on the fly, and training a scene-specific lightweight neural network that learns how light is transported from any scene location to any image pixel. This approach is compatible with non-differentiable production renderers, induces minimal memory requirements during inference, and scales only with resolution and the number of light sources and parameters, but independently of scene and appearance complexity. Our evaluation demonstrates interactive frame rates for relighting (~30-60 Hz) while closely approximating the visual fidelity of ground-truth path tracing. In addition, the differentiable NRP enables fast, gradient-based inverse workflows, allowing artists to efficiently solve for lighting parameters from intuitive image-space edits or generative targets.
comment: Accepted at Computer Graphics Forum (Proc. EGSR 2026)
☆ Validation-Guided Prompt Selection for Differential Morphing Attack Detection WACV 2027
Differential morphing attack detection (D-MAD) compares a trusted live capture with a document image that may contain a facial morph. We investigate whether labelled validation data can improve a frozen-parameter vision-language evaluator through prompt selection rather than model fine-tuning. Version~1 generates 50 candidate prompts and selects the top-ranked instruction using a security-oriented loss that prioritises missed morph attacks (MACER) while accounting for bona fide classification errors (BSCER), response failures, parse failures, and decision inconsistency. Version~2 recursively revises prompts using aggregate validation feedback. On a subject-disjoint test set of 160 pairs spanning LMA, MIPGAN-2, and PIPE, the selected Version~1 prompt achieves a (19.17\%) morph attack classification error rate (MACER), with (0.00\%) observed bona fide sample classification error rate (BSCER), compared with (70.00\%) MACER for a controlled manual prompt. The paired (50.83) percentage-point difference has a (95\%) subject-bootstrap interval of ([43.33,58.33]). Recursive revision does not surpass the best Generation~0 validation loss; the reported revised descendant obtains (31.67\%) test MACER. All 23 attacks missed by Version~1 are landmark-based (LMA) morphs. Overall, the results show that prompt formulation can substantially influence frozen-parameter VLM-based D-MAD performance, while also highlighting the challenge of recursive prompt refinement under a compact 10-subject validation setting.
comment: Accepted in WACV 2027 Workshops
☆ SpotDiff: A Plug-and-Play Framework for Zero-Shot Change Detection in Remote Sensing Imagery
Remote sensing change detection usually relies on large collections of temporally aligned, pixel-level annotations, which limits its scalability across regions, imaging platforms and application domains. Recent zero-shot approaches reduce the need for supervision, but they often localize fine-grained changes poorly because their features are coarse or inconsistent across time. We present SpotDiff, a training-free, plug-and-play framework for zero-shot change detection built by composing complementary vision foundation models. SpotDiff uses DINOv3 for semantically robust features, AnyUp to reconstruct them at pixel resolution, and SAM 3 for text-guided region-of-interest proposals, which an inference-time multi-scale strategy (SWIMS) makes reliable on small objects. Change is then measured by comparing the reconstructed features of the two dates inside the proposed regions, with no learnable components, fine-tuning or change labels. To our knowledge, SpotDiff is the first change detection method to use a universal feature upsampler, which alone adds 13.8 F1 over bilinear upsampling of the same features. On LEVIR-CD, SpotDiff reaches 82.46 F1, 2.06 points above the strongest prior training-free method, and on the multi-class SECOND benchmark it improves building change detection by 10.3 IoU and the six-class average by 2.2 IoU. Qualitative results on PlanetScope (3 m) and Sentinel-2 (10 m) imagery show that the same pipeline transfers across resolutions without modification. Our results indicate that composing modern foundation models is an effective route to scalable, label-free change detection.
☆ UAV-Flow++: A Unified Suite for Benchmarking and Learning Language-Guided UAV Flight
Donglin Yang, Xiangyu Wang, Yantong Zhong, Wenhao Zheng, Zihan Ding, Yue Liao, Hongsheng Li, Ke Xu, Si Liu
Language-guided UAV navigation has largely focused on long-horizon route planning, leaving short-range, object-relative flight maneuvers underexplored. To address this gap, we formulate Flying-on-a-Word (Flow), a task requiring a UAV to convert concise natural-language instruction, egocentric visual observation, and current state dynamics into continuous flight actions. Flow encompasses both primitive motion commands and complex object-interactive behaviors, assessing motion intent comprehension and spatial context grounding. To enable policy learning and closed-loop evaluation, we introduce UAV-Flow, a comprehensive dataset comprising 30,692 expert-piloted real-world trajectories across eight distinct motion categories, alongside 10,109 simulation trajectories in the UAV-Flow-Sim environment equipped with task-specific completion criteria. Furthermore, we develop FlowVLA, which undergoes imitation pretraining on mixed real-sim demonstrations and online post-training via Tail-Aware GRPO (TGRPO). By leveraging group reward distributions, TGRPO effectively reinforces high-reward action sequences. Evaluated on 273 held-out UAV-Flow-Sim episodes, pretrained FlowVLA achieves a mean success rate of 75.58%, outperforming the strongest baseline by 6.72 percentage points, which TGRPO post-training further elevates to 93.16% with superior sample efficiency. Physical UAV experiments demonstrate robust execution of object-relative commands in both indoor and outdoor real-world scenes. Together, Flow establishes a novel benchmark and learning framework for language-conditioned flight maneuvers.
comment: 22 pages, 10 figures
☆ HiVE-VQA: Hierarchical Interaction via Volumetric Embedding for Video Quality Assessment WACV 2027
The proliferation of user-generated video content has made accurate and robust no-reference video quality assessment (NR-VQA) a highly important challenge. Existing research increasingly relies on architecturally complex multi-branch pipelines or computationally intensive fine-tuning of single-branch backbones. In this paper, we investigate whether a single frozen video foundation model can serve as an effective NR-VQA backbone, and propose HiVE-VQA, the first framework (to our knowledge) to apply a JEPA-based video encoder to this task. HiVE-VQA extracts token grids from multiple depths of a pre-trained V-JEPA 2.1 encoder and stacks them into a hierarchical spatio-temporal volume, which a compact 3D convolutional head fuses across its layer, spatial, and temporal axes, combining low-level distortion cues from shallow layers with the semantic and temporal context captured at greater depths. The encoder remains frozen throughout, so only the head is trained: 5.2M parameters (5.7\% of the total), with no auxiliary multi-modal models and an inference cost that is invariant to video resolution. Trained solely on LSVQ and evaluated without any fine-tuning, HiVE-VQA outperforms 17 state-of-the-art (SOTA) methods in SROCC and PLCC on both intra-dataset (LSVQ-test, LSVQ-1080p) and cross-dataset (KoNViD-1k, LIVE-VQC) benchmarks. These results indicate that hierarchical, depth-wise fusion of features from a single frozen video foundation model provides a more effective and efficient basis for NR-VQA than task-specific multi-network fusion. Code and model checkpoints are available at https://github.com/RobHam99/HiVE-VQA.
comment: Accepted to WACV 2027, 11 pages, 3 figures, 6 tables
☆ STeP-C: Real-Time Neural Compression for Distributed Fiber-Optic Sensing at Scale WACV 2027
Distributed fiber-optic sensing (DFOS) technologies such as Distributed Acoustic Sensing (DAS) generate continuous, high-resolution spatio-temporal data at rates exceeding hundreds of megabytes per second, creating severe challenges for real-time storage and transmission. While neural compression methods can achieve strong rate-distortion performance on sensing data, prior approaches either require per-segment retraining incompatible with streaming, or have only been validated offline. None have been demonstrated end-to-end on a continuously streaming DAS acquisition system at scale. We present STeP-C (Spatio-Temporal Predictive Compression), a neural compression framework that decouples spatial encoding, temporal prediction, and entropy-optimized residual coding, trained end-to-end with a rate-distortion objective motivated by conditional entropy reduction. Critically, we identify entropy coding as the dominant bottleneck for streaming deployment. We address it through GPU-side symbol computation, cached probability tables, and parallel GIL-free arithmetic coding, together with FP16 inference and batched entropy coding, enabling a full real-time pipeline built around a GPU ring buffer. We validate this pipeline on a live DAS testbed, achieving a 98.85% file size reduction at 31.04 dB full-band PSNR and an average throughput of 535 MB/s, exceeding the 324 MB/s requirement for 1 kHz, 81 km DAS streams. We further show that STeP-C substantially outperforms conventional codecs at practical streaming bitrates, can be applied to unseen channel ranges within the same acquisition system without retraining, and largely maintains downstream detector consistency, with 96.62% high-confidence class agreement and 100% Jaccard similarity on detected event types.
comment: Accepted to WACV 2027 (Applications Track). 10 pages, 9 figures, 3 tables
☆ Spatiotemporal learning improves quantification of pore-space evolution and fluid displacement
Pore-space evolution and fluid displacement influence how rocks transmit and store fluids, affecting geological carbon storage, underground hydrogen storage and groundwater contaminant transport. A single time-resolved micro-CT experiment may produce tens to hundreds of three-dimensional images, and segmentation is required to quantify physical properties and run image-based simulations. Traditional methods such as thresholding perform poorly on noisy, low-contrast images and need manual adjustment per image; 3D deep learning networks segment each image without temporal information, so their errors vary between images and could be mistaken for physical change. Here we introduce a transformer-based 4D segmentation framework combining spatial and temporal information. It uses sequential images to resolve ambiguous phase assignments and separate physical change from segmentation error. The method is tested on images of mineral dissolution: it measures pore-volume growth and rock boundaries more accurately than an equivalent 3D model that ignores temporal information. It also captures the increase in permeability verified against reference segmentations and flow simulations. The advantage transfers to another dissolution sample without retraining. On two other datasets of fluid displacement, it reduces spurious changes in predicted rock and porosity fluctuations and accurately identifies filling events. Spatiotemporal segmentation turns sequences of hundreds of three-dimensional images into more reliable quantitative measurements of pore-space evolution and fluid displacement without per-image parameter adjustment to support the analysis of time-resolved CT imaging.
comment: 48 pages, 9 figures, 18 tables, including Supplementary Information
☆ Predicting Ethanol Drinking Status from Nonhuman Primate Liver Histology: A Proof of Concept for Parameter-efficient Fine-tuning in Few-Shot Vision-Language Modeling
Combining vision-language models (VLMs) with parameter-efficient fine-tuning (PEFT) offers a promising approach to few-shot whole-slide image (WSI) classification in computational pathology. This study evaluates this approach for predicting ethanol drinking status from nonhuman primate liver histology. Adapters are evaluated against non-PEFT configurations on a subset of the Monkey Alcohol Tissue Research Resource (MATRR). All evaluated PEFT configurations achieve higher mean balanced accuracy than the non-PEFT baselines. For example, in the four-shot setting, attention-based multiple instance learning (ABMIL) alone achieves 43.5% mean balanced accuracy, compared with 78.1% after adding an adapter. Qualitative attention analysis shows that textual pathway adaptations produce localized attention heatmaps, while visual pathway adaptations produce more diffuse ones. A pathologist's review identified similar histological features across adaptations.
comment: 3 pages, 2 figures. Accepted for publication in the proceedings of IEEE BIBM 2026. The code is publicly available: https://github.com/Advanced-Vision-and-Learning-Lab/ALD_VLM
☆ TuMAP: A Dual-View, Morphotype-Aware Pathomics Pipeline for Quantifying Renal Tubular Remodeling
Ruijie Wu, Yuechen Yang, Yanfan Zhu, Yu Wang, Shilin Zhao, Ming-Tsun Tsai, Szu-Yuan Li, Shuo-Ming Ou, Tz-Heng Chen, Haichun Yang, Yuankai Huo
Quantifying disease-associated renal tubular remodeling is challenging because tubular appearances are heterogeneous. We introduce TuMAP (Tubular Morphotype-Aware Pathomics), a pipeline combining (1) dual-view foundation-model representations and provisional morphotype annotation, (2) compositional and within-morphotype pathomics, and (3) patient-level statistical inference. From 456,158 tubules in 358 PAS-stained kidney biopsy slides, TuMAP identified eight image-based morphotypes, seven of which showed stable subsampling assignments. Across 219 patients in six diagnostic groups, diabetic nephropathy (DN) and tubulointerstitial nephritis/acute tubular injury (TIN/ATI) each differed from minimal change disease in seven morphotype proportions, while IgA nephropathy differed in six. DN showed reduced nuclear solidity and increased eccentricity across all eight morphotypes. Area-normalized nuclear density exposed additional within-morphotype differences. TuMAP provides a framework for separating changes in tubular morphotype composition from phenotypic differences within morphotypes.
comment: 5 pages, 3 figures, 1 table. Submitted to IEEE ISBI 2027
☆ Joint Null-Space Projection for Data-Free Model Merging
Merging independently fine-tuned checkpoints into a single multi-task model requires no data, no labels, and no gradient steps. In practice the task vectors of different tasks interfere, and the standard remedies--averaging, magnitude pruning, sign election--operate entry by entry, ignoring that each delta induces a low-dimensional update on the hidden features it acts upon. We recast merging as a subspace interference problem. For every two-dimensional weight tensor we take the top-r right-singular subspace of each task's delta--the input directions along which that task moves the representation most--and remove from each task vector the component lying in the union of the other tasks' subspaces. The resulting Joint Null-Space Projection (JNSP) is closed-form and order-independent, needs no forward pass, and is defined entirely by the checkpoint set. On eight CLIP image-classification tasks JNSP merges ViT-B/16 experts to 89.45% average accuracy without using a single image, 11.88 points above the strongest parameter-wise training-free baseline (TIES-Merging, 77.57%), 14.22 above Task Arithmetic, and 0.98 above WUDI's data-free consensus merge (88.47%). The gain holds across three backbones on the independently trained TALL release (+10.8, +12.6, +9.1 over TIES for ViT-B/16, ViT-B/32, ViT-L/14) and replicates there at 89.20%; controls confirm the source--a single shared subspace instead of per-task ones loses 2.9 points, and a sequential variant collapses to 69.6%. On RoBERTa-large the same projection underperforms plain task arithmetic, which we report rather than hide. Code is available at https://github.com.
☆ SALT: State-Aligned Low-Rank Transfer for Single-Temporal Cross-Dataset Structural Localization in Remote Sensing WACV 2027
Aggelos Psiris, Georgios Angelidis, Giorgos Sfikas, Maria Vakalopoulou, Thomas Lagkas, Panagiotis Sarigiannidis, Georgios Th. Papadopoulos
Change detection in remote sensing conventionally relies on spatially aligned bi-temporal imagery, which may be unavailable or unreliable in practical scenarios. Recent single-temporal approaches address this limitation by learning structural evidence, such as building regions, from a labeled source dataset and transferring it to unseen target imagery using only the current image. However, adaptation to the source dataset can over-specialize pretrained features and reduce their cross-dataset generalization, a risk that this paper argues is higher when the feature adaptation is parameterized independently of the states used for prediction. To address this issue, the current study introduces SALT, a State-Aligned Low-Rank Transfer framework. SALT keeps a DINOv3 backbone frozen and learns negative and positive structural states that both score each patch and, through State-Aligned Low-Rank Adaptation (SALA), span the rank-2 subspace in which patch features are corrected. Directional State Alignment supervises these corrections to raise the evidence of positive patches and lower that of negative ones, while Geometry Preservation limits changes to the pairwise relations among pretrained patch features. Experiments on LEVIR-CD and the WHU aerial building dataset show that, at the pixel level and in both transfer directions, SALT outperforms four baselines reproduced under a common source-only protocol, improving over the strongest reproduced baselines by 9.45 and 5.79 points in pixel-level AP and by 27.69 and 14.25 points in IoU (LEVIR-CD to WHU and WHU to LEVIR-CD, respectively), with only 0.068M trainable parameters and 40.2 ms per image. In order to facilitate reproducibility and further research, the source code of SALT is available at https://github.com/gthpapadopoulos/SALT
comment: Accepted at the Third Workshop on Computer Vision for Geospatial Image Analysis (GeoCV), IEEE/CVF WACV 2027
☆ Ropedia Xperience 2: A Gold Standard for Scaling Physical AI Environments and Behaviors
Hao Li, Fangzhou Hong, Zhaoxi Chen, Hui En Pang, Kairui Hu, Mingze Sun, Runmao Yao, Xinhao Ji, Xiyuan Zheng, Xuanli Chen, Yalun Dai, Yap Kim Hui, Dingwen Zhang, Yuanyuan Gao, Ziang Cao, Cheng Huang Yean, Jackson Peng, Rui Duan, Yaobo Liu, Zetong Wang, Ziwei Liu
Large foundation models such as GPT-6 Astra and Claude Fable 5.1 have recently been shown to control robots directly from visual observations. They nevertheless remain brittle on long-horizon, contact-rich tasks in real-world environments. Further progress depends on two complementary resources: demonstrations of human behavior that supervise how tasks are performed, and interactive environments in which policies can be rolled out, evaluated, and improved in closed loop. Language models have advanced along well-characterized scaling laws; the corresponding question for physical AI is how capability scales with environments and behaviors, which first requires identifying which data to scale and how to produce and reuse such data continuously. To address this question, we introduce Ropedia Xperience 2, comprising 1,589 released metric-scale 3D scans and a collection of 4,230 egocentric recordings of 74 tasks (131 hours). We further show that each capture can be repurposed for robot learning: scenes are converted into physically interactive simulation environments, human motion is retargeted to whole-body robot trajectories, and additional robot behaviors are obtained through virtual teleoperation and automated long-horizon trajectory generation. We propose this joint capture of environments and behaviors as a gold standard for physical AI data collection, laying a data foundation for establishing scaling laws over environments and behaviors.
comment: https://ropedia.com/xperience/2-research
☆ AURA 2.0: From User-Driven Responses to Model-Initiated Conversations in Streaming Video
Shuhan Li, Huankang Guan, Xintong Guo, Yang Bo, Yuyang He, Dunyuan Xu, Jingyan Chen, Fang Liu, Tsz Kin Wan, Jinpeng Chen, Zhenqi He, Yuxuan Hu, Wenbo Li, Han Gao, Qijun Li, Long Chen, Hongsheng Li, Rui Liu
Most AI assistants respond only when prompted, remaining silent even when events in the video stream offer meaningful opportunities for conversation. To move beyond purely user-driven interaction, we present AURA 2.0, a streaming video large language model that introduces three core upgrades to AURA. The key advance is Model-Initiated Conversation (MIC), which enables the model to identify topics worth discussing in the evolving visual stream and initiate conversations at appropriate moments without a user query. Stream-Aware Tool Calling (SATC) enables the model to invoke external tools using the current visual and conversational context and incorporate returned information into the ongoing interaction. Interactive Visual Grounding (IVG) supports visual grounding in both user inputs and model outputs. The model can identify objects marked by users on video frames and produce frame-aligned bounding boxes to link its responses to specific objects and regions. To develop these capabilities, we construct a dedicated streaming interaction dataset and adopt a three-stage training scheme to integrate the new behaviors with AURA's existing streaming interaction abilities. We further incorporate Long-Horizon Interaction Memory (LHIM) to maintain contextual continuity over extended video streams. Evaluations on public streaming benchmarks show that AURA 2.0 achieves leading performance among open-source streaming models, while targeted evaluations further validate its new interaction capabilities. We open-source the model weights and inference code.
☆ ElasticDino: Dense Semantic Features Using Elastic Deformations
Vision transformers (ViTs) serve as the backbone for countless vision tasks, including dense prediction tasks, such as segmentation and depth estimation, even though the quadratic complexity of self-attention severely limits feature resolution and induces spatial artifacts. In this work, we introduce ElasticDino, a lightweight adapter which produces high-resolution feature maps, adhering to the fine details of the input image. As a key insight, we observe the consistency of artifacts across channels, and simultaneously resolve them to match the input image structure. Using a slim architecture based on spatial deformations, ElasticDino produces pixel-aligned, high-resolution feature maps efficiently. Leveraging deformations allows strictly preserving sharp semantics, and thereby avoiding the smoothness biases inherent to alternative approaches. Due to the absence of ground truth, we employ pixel-level semantic segmentation and depth estimation with linear heads to predict a set of downstream dense tasks, guiding the deformation process without leaking task-specific semantics into the model. We demonstrate superior accuracy over current methods on a range of popular dense task benchmarks, with better computational performance.
☆ A Disease-specific Retinal Aging Biomarker for Predicting Progression to Late Age-related Macular Degeneration
Age-related macular degeneration (AMD) is a leading cause of irreversible vision loss in older adults. While chronological age is a primary risk factor, it inadequately reflects individual variations in retinal aging, limiting effective AMD risk stratification and timely intervention. Here, we propose and validate fundus-based AMD age (FAA), which quantifies disease-specific retinal aging in AMD and captures accelerated AMD-related retinal aging, thereby improving prognostic precision over chronological age. Trained on 36,782 images from 2,192 participants in the Age-Related Eye Disease Study (AREDS) dataset and validated across AREDS and AREDS2 cohorts, our model demonstrates significant improvements over traditional chronological age-based approaches. Specifically, integrating the proposed FAA into Cox proportional-hazards models significantly enhances prediction performance. FAA improved the C-index from 0.836 to 0.888 and reduced the 5-year Brier score from 0.050 to 0.046 in AREDS, with consistent gains observed in AREDS2 (C-index 0.715 to 0.756; 2-year Brier 0.078 to 0.076). FAA showed improved prognostic performance across the two datasets and under different phenotype definitions (reading center-defined vs. specialist-defined grading), supporting its potential utility for AMD risk stratification. These findings highlight the advantage of incorporating FAA into AMD risk prediction frameworks, supporting further evaluation of FAA for AMD risk stratification.
comment: 41 pages, 13 figures (including supplementary material)
☆ Report of the 8th LSVOS Challenge: Mobile Unified Multimodal Understanding ECCV 2026
Chang Liu, Henghui Ding, Tong Yang, Dayoung Kil, Seong-heum Kim, Qilin Xiang, Ronghao Zheng, Ziyang Zhang, Shanlin Jiang, Ruohan Shi, Qilin Fan, Chengfeng Qiu, Kaifeng Wei
This report summarizes the Mobile Unified Multimodal Understanding (MUMU) track of the 8th Large-scale Video Object Segmentation (LSVOS) Challenge, held in conjunction with ECCV 2026. MUMU asks for a single unified model that jointly performs multi-concept image tagging over quality, scene, and event vocabularies, open-vocabulary object detection, and concise image captioning, within a budget of 0.5B parameters and 8 GB of peak inference memory. Submissions are ranked by a weighted task score scaled by an efficiency factor. We describe the tasks, rules, and evaluation protocol, report the final results, and review the methods of the top three teams. All three leading solutions build on the compact Florence-2-base model, but extend it in different ways: lightweight tagging heads with cross-task inference refinement, a shared low-rank adapter trained on coverage-aware multi-source data with task-dependent decoding, and a training-free pipeline that maps captions to tags and applies detection test-time augmentation. The results show that a compact prompt-based vision-language model can serve all three outputs when supervision, decoding, and cross-task cues are designed for each task, while open-vocabulary detection on novel categories remains the main bottleneck.
comment: 13 pages, 3 figures, 3 tables. Report of the MUMU track of the 8th LSVOS Challenge, held in conjunction with ECCV 2026. Project page: https://lsvos.github.io/mumu
☆ Automated PD-L1 Tumor Proportion Scoring with Uncertainty-Guided Triage in Non-Small Cell Lung Cancer
Kesi Xu, Ayesha Azam, Tanvier Kaur, Monika Kondratowicz, Kinza Asim, Brinder Chohan, Paapa Appiah-Odame, Zhao Chen, David Snead, Nasir Rajpoot
Programmed death-ligand 1 (PD-L1) tumor proportion score (TPS) informs immunotherapy selection in non-small cell lung cancer. Near the 1% and 50% cutoffs, the score should also indicate when closer review is warranted. We designed a three-stage framework to follow the sequence of TPS assessment. Stage 1 identifies tumor regions and excludes macrophage-rich patches. Stage 2, a dual-branch cell counting network (D2Count-Net), counts PD-L1-positive and PD-L1-negative tumor cells in these regions to calculate TPS. Stage 3, TPS-Guard, an uncertainty-guided triage mechanism, reports model uncertainty as the disagreement among independently trained models, refers the most uncertain slides, and flags scores near the 1% or 50% cutoff when model estimates disagree across that cutoff. Flags state the reason for referral or alert the pathologist to possible scoring errors on auto-reported slides. In a retrospective, single-center study, 154 whole-slide images were scored by three pathologists, and the model and each pathologist were compared in turn with the mean of the other two. On the 114 slides unseen in training, model mean absolute error (MAE) was 6.96 percentage points, within the range of the three pathologists (4.24 to 8.29), and model uncertainty correlated with absolute TPS error (Spearman rho = 0.77) and inter-pathologist disagreement (rho = 0.71). After referral of the 15% most uncertain slides, model and pathologist MAE both decreased. Applied without retraining to an external, publicly available dataset, TPS-Guard still directed review to higher-error cases, and referring more uncertain cases raised sensitivity at the 50% cutoff. TPS-Guard provides reasons for reviewing individual automated scores and identifies slides that also produce greater disagreement among pathologists. Prospective reader studies should assess TPS-Guard's effect on TPS reporting.
comment: 55 pages (29-page main text with 5 figures and 3 tables, followed by 26 pages of supplementary material)
☆ TokenStream: Multi-Subject Personalized Video Generation with Token-wise LoRA and Guided Attention WACV 2027
We introduce TokenStream, an optimization-based framework that composes independently learned subject identities in text-to-video generation without a reference-video training dataset or joint optimization of each new subject combination. Learning from a single reference image makes subject appearance difficult to separate from static-image bias, while merging independently trained adapters can mix identities. Token-wise LoRA (ToRA) restricts low-rank updates to the cross-attention value projections of subject-associated text tokens, providing a separate adaptation path for each subject. A T2V-guided stream copies queries and keys from the frozen base model during early denoising to reuse its layout and motion priors while retaining personalized values. We also introduce Multi-DB, a 30-subject, 122-prompt benchmark with appearance-free prompts and varied interactions. On Multi-DB at 1.3B, TokenStream improves identity preservation from 0.422 for DreamBooth to 0.467 and textual alignment from 0.652 to 0.688.
comment: Accepted to WACV 2027
☆ Raysemble: Metric 3D Bone Reconstruction from a Single X-ray
Raysemble reconstructs eight lumbar and pelvic bones from a single X-ray by separating shape and placement. Each 3D query samples regional and global features at its image projection. A shared decoder predicts normalized bone shapes. Predicted depth and camera geometry determine 3D centers and scales for resampling these shapes onto a common grid in millimeters. Evaluation uses 23 held-out patients with 43 CT scans, each generating one simulated anteroposterior radiograph. Without rigid alignment to CT,Raysemble leads 15 evaluated methods, averaging 65.24% Dice, 9.09 mm HD95, and 3.32 mm ASSD over three random seeds. With identical inputs, it improves Dice over VolumeNeRF by 5.86 percentage points.
☆ Visual Cipher Bench: Benchmarking LLM Agents on Long-Horizon Visual Reasoning
Deyao Hong, Jiayi Jiang, Yuchun Chen, Yixuan Shi, Wenyi Li, Zheng Jiang, Chengxing Xie, Mingyu Su, Qingle Liu, Mingju Gao, Yizhe Chi, Youjie Zheng, Calvin Xiao, Kaisen Yang, Huan-ang Gao
While language-model agents excel at long-horizon coding and terminal tasks, this success relies on two conditions: explicit text specifications and immediate execution feedback. However, neither condition holds in real-world problems where critical information exists only in images, leaving it unclear whether long-horizon capability transfers from text to visual reasoning. In such visual settings, because rules and clues resist text transcription and lack test-case feedback, agents must repeatedly inspect images, simulate spatial transformations, and autonomously cross-check clues. Despite this challenge, existing evaluations fail to capture such capability: visual reasoning benchmarks largely reduce tasks to single-turn, short-chain question answering, whereas agent benchmarks focus on text-based environments or interactive states that permit trial-and-error. To bridge this gap, we introduce Visual Cipher Bench, a collection of 40 original visual puzzles across four categories: spatial manipulation and simulation, logical constraints and planning, rule induction, and knowledge-based decoding. Each puzzle provides only one to three black-and-white line drawings and a minimal submission instruction, requiring agents to discover the underlying solving mechanism entirely from the images. Across 29 reasoning-effort configurations of eight frontier models, even the strongest model, GPT-6 Astra (max), achieves only 26.7% accuracy, while five of the eight models score below 1%. Yet on a ten-task subset, providing a single non-spoiler mechanism hint boosts accuracy from 7.8% to 55.6%. This sharp contrast shows that the primary bottleneck of current agents lies not in executing multi-step plans, but in autonomously discovering rules from visual evidence. Visual Cipher Bench thus provides a rigorous benchmark for measuring long-horizon visual reasoning in agents.
☆ Kosmo-GFM: A Production-Ready Geometry Foundation Model
Sheng Tang, Shengzhe Wei, Tiancheng Chen, Wenhua Jin, Xinyang Lu, Tinghao Jiang, Shuoxuan Feng, Yupeng Liu, Yu Zhang, Xingshuo Chang, Keren Zhou, Yuhan Xiong, Jiahao Li, Haoming Liu, Jituo Tian, Hui Liang, Chaoqun Tang, Feng Tian, Wentao Ouyang, Kaojin Zhu, Xiaolin Chai, Peihuang Fu, Miao Zheng, Yunbo Hu, Feixue Cai, Zizhen Zou, Weigang Wang, Yexuan Yang, Haoyu Feng, Weiqi Zhang, Juntong Fang, Zesong Li
We present Kosmo-GFM, a scene reconstruction system that integrates a geometry foundation model (GFM), a pose optimization engine (PoseEngine), and reconstruction-derived supervision from KosmoWorld. GFM predicts camera parameters, depth, cross-view correspondences, and dynamic 3D points from images and videos. PoseEngine combines these predictions with geometric optimization to recover scene-level camera poses and geometry for downstream 3D and 4D Gaussian Splatting. KosmoWorld generates images and geometric supervision from reconstructed real-world scenes to improve GFM's geometric predictions. On the Sparse and Medium subsets of SpatialBench, GFM achieves the lowest absolute relative depth error and the highest relative-pose AUC at 5 degrees among the compared methods. It also achieves the lowest mean endpoint errors for dynamic points and scene flow across the five evaluated datasets. Across 34 scenes completed by all compared reconstruction systems, PoseEngine achieves the highest scene-averaged pose AUC at thresholds of 1, 3, and 5 degrees, with a cumulative runtime of 14.07 hours, approximately 20% of GLUEMAP's 69.67 hours. PoseEngine Flash provides a further accuracy-efficiency trade-off, with a measured mixed-precision runtime of 2.62 hours, approximately 4% of GLUEMAP's runtime. Separate scaling experiments show generally favorable trends with increased data, model computation, and training budgets. These results support the use of learned geometric priors within a scene-level reconstruction pipeline.
comment: Project page: https://kosmoresearch.github.io/Kosmo-GFM/
☆ EG360: Efficient and Generalizable $360^\circ$ Depth Estimation via Adaptive Panorama-Perspective Distillation and Spherical Cross-Level Feature Fusion
Generalized panorama depth estimation has witnessed great advances in recent years. These methods typically achieve robust zero-shot generalization performance using large pretrained models, but their high computational cost makes real-time inference on embedded platforms challenging. In contrast, previous lightweight panorama models have mostly been developed for single-domain evaluations and cannot generalize well due to domain shifts. To reduce the gap between these two types of models, we introduce EG360, an efficient and generalizable panorama depth estimation model. Specifically, building on a lightweight depth model, we design an Adaptive Panorama-Perspective Distillation (APPD) strategy guided by scene geometry, along with a Spherical Cross-Level Feature Fusion (SCFF) module for global semantic perception. As a result, with only 6.593M parameters after reparameterization, EG360 achieves strong zero-shot performance on real-world benchmarks and runs at 35 FPS on a Jetson Orin NX 16GB in MAXN mode, whereas a previous ViT-small model runs at less than 2 FPS. Our code will also be made public for future research.
☆ Evaluating Video LLMs' Understanding of Corner Cases in Autonomous Driving BMVC 2026
Gabriel Manalu, Minh Son Hoang, Hyeontaek Hwang, Jaesung Lee, Hyojin Choi, Tran Tuan Phong, Seon Kwon Kim, Saad Wazir, Daeyoung Kim
Natural language explanations promised by modern Video Large Language Models (Video LLMs) offer autonomous driving systems a path toward detailed scene description and interpretable decision making on the road. While recent benchmarks have begun assessing these models on driving scenes, they mainly target routine traffic conditions and general comprehension tasks based on existing driving datasets and therefore fail to evaluate model performance on the rarest and most safety-critical driving events, also known as "corner cases". To provide a comprehensive evaluation of Video LLMs on precisely these scenarios, we introduce DashBench, a benchmark testing 25 Video LLMs on 1,709 carefully curated driving videos spanning not only collision-related corner cases but also rare non-collision ones. Through extensive prompts and open-ended question answering, we expose a critical capability gap: all evaluated models perform substantially below the threshold required for safety-critical driving applications, frequently misjudging or missing dangerous events on the road ahead and hallucinating non-existent agents or actions in their explanations. We believe DashBench, combining an extensive dataset with a rigorous evaluation toolkit and a detailed diagnosis of current model limitations, drives progress toward safer, more explainable autonomous driving systems. The code and data will be shared at: cornercase-dashbench.github.io
comment: Accepted at BMVC 2026
☆ Efficient Periocular Recognition through Model Scaling and Cross Spectral VIS NIR Verification
Periocular recognition offers an efficient alternative to full-face verification when facial regions are occluded or image resolution is limited. We present AdaEye, a systematic study of input and backbone scaling for periocular verification in visible-spectrum (VIS) and cross-spectral (VIS/NIR) settings. Using an AdaFace-style framework, we compare full-face, masked-face, and dedicated 80x40 periocular inputs while varying backbone width and embedding dimensionality. Across four VIS/VIS and VIS/NIR protocols, we assess verification accuracy, computational cost, inference throughput, training memory, template size, and training stability. Direct periocular processing reduces recognition cost from 6.30 to 1.64 GMACs per image and increases measured throughput by approximately threefold. Its effectiveness depends on the evaluation protocol: periocular models perform better on native UFPR periocular images, whereas face-based protocols favor retaining more input information. Moderate backbone width reductions maintain comparable mean verification performance on CASIA and ND-NIVL while reducing computation. Embedding compression decreases template-storage requirements, primarily benefiting deployment. Multi-seed experiments reveal that some configurations require numerical stabilization, particularly when embedding and backbone reductions are combined. These findings support selecting compact periocular architectures according to the target protocol and deployment constraints, balancing accuracy, efficiency, storage, and robustness.
comment: Munuscript In revision Process, abstract limited to 1920 ch
☆ AgriSense: Vision-Language Decision Support and Robotic Intervention in a Virtual Greenhouse Copy title
Autonomous greenhouse management increasingly requires robotic systems that integrate perception, navigation, environmental regulation, and plant-level intervention. This paper presents AgriSense, a framework combining an Isaac Sim virtual greenhouse, a quadruped mobile manipulator, RGB and LiDAR-inertial sensing, environmental control, and supervised vision-language decision support. Frozen DINOv2 and BERT encoders, together with structured environmental inputs, support joint prediction of twelve environmental references, a target zone, and an intervention flag. Checkpoint replay reproduces 41/43 zone decisions. Across five seeds on a source-grouped test set, the full, image-removed, and language-removed configurations each achieve 43/44 correct zone decisions, providing no evidence of incremental multimodal zone-accuracy improvement. The full model achieves mean absolute errors of 0.249 degrees C for temperature, 2.030 percentage points for humidity, and 0.997 percentage points for soil moisture. Sensor-only inputs match full-model zone accuracy, while language-only inputs achieve 68.64%. External real-image benchmarks reveal distribution-shift limitations. Five offline Point-LIO routes yield rigidly aligned trajectory errors of 5.22-17.32 mm. Of 40 virtual-arm targets, 32 satisfy geometric acceptance criteria and eight are rejected at boundaries. Six long-horizon climate cases and fault protocols characterize integration limits. Navigation and integrated treatment rely on simulator-state inputs; physical greenhouse performance and agronomic treatment efficacy remain unvalidated. The contribution is an evaluated decision-support and robotic-intervention framework for a virtual greenhouse.
☆ ORCA:Orthogonal Residual Consensus Alignment for Multi-View Clustering
Deep multi-view clustering often suffers from redundant information accumulation during fusion and indiscriminate alignment across representation levels. We propose Orthogonal Residual Consensus Alignment (ORCA) to address these challenges. Its semantic incremental recurrent fusion module extracts cross-view orthogonal residuals as non-redundant increments and adaptively weights recurrent states through divergence-guided gating. Its consensus-calibrated hierarchical contrastive alignment module uses the global consensus to weight contrastive alignment of view-level, pairwise, and recurrent representations. Together, these modules balance view-specific structure with cross-view consistency. Experiments on five benchmarks against fifteen competing methods demonstrate ORCA's effectiveness, including a 5.25 percent accuracy improvement over the strongest baseline on 100leaves. Ablation studies validate both modules. Code is available at https://github.com/XiaojianDing/2026-NeurIPS-ORCA.
comment: 24 pages,5 figures,2 tables
☆ PhysLoop: Harnessing LLMs' Physics Awareness for Self-Improving Video Generation Models
Yuxuan Chen, Xiaojie Li, Hao Chen, Juncheng Ma, Kewei Zhang, Zhongjing Du, Guoqiang Wei, Fangfang Wang, Zhen Dong, Zhiyang Dou, Liming Zhao, Yan Zeng, Jianwei Zhang, Daquan Zhou
Video generation models produce visually realistic videos but still struggle with physical dynamics and causal consistency. Learning from generated videos requires fine-grained physical supervision, yet reliable quality judgments, correction pairs, and controlled variations are difficult to obtain. Large language models (LLMs) possess physical knowledge, but struggle to reliably guide video generation models in producing their training videos. We present PhysLoop, a unified harness designed to support physically grounded self-improvement in video generation models. The harness guides and constrains LLMs throughout video generation, selection, and correction, distilling their physical knowledge into physically consistent videos. The framework supports controllable training-data construction and inference-time video repair. By identifying and correcting failures, PhysLoop marks a key milestone toward recursive self-improvement (RSI) in video generation models with limited human intervention. We release 100K videos spanning diverse, realistic indoor and outdoor scenes, along with generation tools and two evaluation datasets: PhysLoop-Gen and PhysLoop-Repair. We evaluate controllable generation and localized repair on these datasets, respectively, and assess physical behavior on the external Physics-IQ Verified benchmark.
☆ Fog-FWL: A Large-Scale Full-Waveform LiDAR Dataset for Recovering Fog-Hidden Objects
Kazuma Ikeda, Kotaro Oishi, Ryosei Hara, Ryo Yoshida, Takahiro Kado, Ibuki Fujioka, Taro Beppu, Mariko Isogawa, Kentaro Yoshioka
LiDAR is a core sensor for autonomous driving, yet it is highly vulnerable to fog. Fog scattering produces strong backscatter near the sensor and attenuates reflections from objects behind it, so weak object returns are often discarded by the peak detection that converts received signals into point clouds. Full-waveform LiDAR (FWL), which records the received optical signal before peak detection, retains these weak returns, but the lack of suitable data has prevented research on recovering fog-hidden objects from FWL. We present Fog-FWL, a dataset foundation for this task that consists of a real-world benchmark, Fog-FWL-Real, and a large-scale synthetic training set, Fog-FWL-Sim. Fog-FWL-Real contains 7,527 real foggy FWL frames from 8 scenes and 58 sequences, which is 579x larger than the only existing real foggy FWL data, together with histogram-level Object/Fog/Noise labels, 10,719 3D bounding boxes, and three difficulty levels defined by the attenuation of object reflections. Fog-FWL-Sim contains 12,000 frames generated by our physics-based simulator, Fog-FWL-Simulator, which integrates spatially heterogeneous fog density fields, time-resolved light transport including multiple scattering, and a photon-counting SPAD receiver model. The two subsets share the same sensor format and annotation protocol, enabling a strict sim-to-real evaluation in which no real waveforms are used for training. Our analysis shows that, under the hardest condition, object returns are included in the three strongest peaks for only 34.9% of object rays, underscoring the need for waveform-level recovery. Fog-FWL-Real, Fog-FWL-Sim, and the Fog-FWL-Simulator will be publicly released to facilitate research on fog-robust LiDAR perception.
comment: 6 pages, 1 figure, 3 tables
☆ From Foundation VLM to Versatile Driving VLM: A Triplet Incremental Training Approach
Adapting general Vision-Language Models (VLMs) to Autonomous Driving (AD) is thought to require heavy 3D spatial priors or intrusive architectural overhauls to compensate for a lack of spatial grounding. In this work, we present a counterintuitive pilot study: off-the-shelf general VLMs already exhibit remarkable adaptability across diverse driving tasks, while precise metric depth remains the primary missing link, a bottleneck readily addressable via onboard sensors and depth foundation models. To this end, we introduce the Triplet Incremental Training Framework, a non-intrusive paradigm that seamlessly instills 3D geometric awareness into a general VLM backbone while fully preserving its open-world capabilities. Driven by automatically constructed (image, depth, QA pair) triplets, our framework operates via a unified tri-stream training pipeline: an unfrozen primary VLM stream to adapt to driving tasks, an auxiliary depth stream that guides the visual backbone to implicitly imagine 3D scene geometry via explicit geometric constraints, and a frozen twin VLM stream employing logit-level KL distillation to prevent semantic drift. Crucially, the auxiliary depth branch is discarded during deployment, ensuring the enhanced geometric awareness comes with zero inference overhead. We evaluate our framework across four major benchmarks targeting four diverse capabilities spanning 3D spatial perception, driving-oriented language reasoning, end-to-end planning, and general multimodal understanding. Our approach outperforms state-of-the-art driving VLMs, rivals domain-specific expert models in 3D perception, and successfully retains its original general reasoning capabilities. This demonstrates that Triplet Incremental Training offers a unified, deployment-ready paradigm for transforming general foundation models into robust driving intelligences.
comment: 13 pages, 3 figures, 7 tables. Accepted at the 10th Conference on Robot Learning (CoRL 2026)
☆ What Limits Thermal Infrared Drone Detection? A Cross-Dataset Benchmark Analysis WACV
Thermal infrared (IR) imaging enables drone detection with reduced dependence on ambient illumination, supporting continuous surveillance beyond the limitations of visible-spectrum sensing. However, substantial and often undocumented variability in camera systems, processing pipelines, and annotation practices across public IR drone datasets complicates the assessment of detector performance and cross-domain generalization. To examine these sources of variability and their implications for drone detection, we systematically characterize existing IR datasets and establish an artifact taxonomy reflecting their origins across the acquisition and processing pipeline, encompassing optical and sensor effects, tone mapping and image enhancement, and recording-related artifacts. Focusing on long-wave IR data, we investigate the impact of training data composition and model characteristics on YOLO-based detection performance, generalization, and condition-dependent failure modes. Our results demonstrate that detection performance is governed more strongly by training data composition than by YOLO generation or model capacity, while cross-dataset generalization exhibits pronounced asymmetries. Even the best-performing configuration fails to detect over 40\% of drones, with elevated miss rates for very small targets, cluttered urban backgrounds, and thermal crossover, highlighting the shortcomings of existing datasets.
comment: Accepted at 2027 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW), Workshop "Real World Surveillance: Applications and Challenges, 7th"
☆ Sprint3D: Accelerating End-to-End 3D Asset Generation through Pipeline Optimization
Sho Kuno, Ruihan Yu, Yonghao Yu, Guixu Lin, Lian Fu, Muyao Niu, Yu-Ju Tsai, Zheng-Hui Huang, Yi-Chuan Huang, Siqi Yang, Fengbo Lan, Erwin Wu, Kaipeng Zhang, Zhixiang Wang
Image-to-3D generative models now produce asset-quality textured meshes, but at minutes per asset they remain slow to iterate with and costly to deploy. Prior work has mostly accelerated the sampling of the networks, and where it also optimizes the mesh processing, it trains its own generators. We instead start from a publicly released pipeline and ask how much of its end-to-end time can be saved. Profiling the public Pixal3D pipeline shows that its latency is dominated by what happens after the networks: the four rectified flows take 14% of the wall-clock time and mesh post-processing takes 85%, most of it in CPU-based UV unwrapping. We therefore first rebuild the mesh post-processing on the GPU and take it off the critical path, then cut the per-evaluation overhead of the flows and distill them from 76 to 8 network evaluations. The resulting pipeline, Sprint3D, reduces the median time per asset from 133 s to 4.05 s on one H200 GPU at the same export settings (33$\times$). At the same time, on 2,653 held-out Toys4K objects it improves every input-view normal metric of the Pixal3D protocol, with a slightly lower mean silhouette IoU, lower point-cloud alignment scores and less fine texture detail. The rebuilt post-processing alone also speeds up TRELLIS.2, which uses the same export code, by 6.7$\times$ without a detectable change in the evaluated metrics.
comment: Project page: https://alaya-lab.github.io/sprint3d/ Code: https://github.com/AlayaLab/sprint3d
☆ MuViRL: Multimodal View-Invariant Visual Speech Representation Learning via Calibrated View Aggregation
Multimodal representation learning combines complementary signals to capture shared semantic information. For dynamic scenes, these representations need to remain consistent across views while preserving distinctions between temporal states. In visual speech representation learning, view variation changes the visibility and reliability of articulation cues, making it difficult to learn such representations from synchronized audio and video. To address this challenge, we propose MuViRL, a self-supervised framework for multimodal view-invariant visual speech representation learning via calibrated view aggregation. The framework combines synchronized 3D Gaussian view synthesis with multimodal self-supervision to learn consistent representations from real and synthesized data. A multimodal decision model learns relative view utility from teacher prediction losses, and its calibrated scores guide the aggregation of complementary information. Aggregation-to-Single Distillation transfers the aggregated predictions to a single branch. Real-synthetic alignment matches synchronized windows across views while preserving their temporal relations. Experiments on real and synthesized multiview data demonstrate that MuViRL improves multimodal representation consistency and visual speech recognition performance across views.
comment: 30 pages, 7 figures
☆ Learning the Missing Derivative: Neural Multiple-Scattering Operators for Inverse Volume Rendering
Reconstructing optically thick media from photographs requires modeling multiple scattering and its response to changes in extinction. Frozen-source tomography discards this derivative; differentiable path tracing estimates it through expensive, noisy sampling. We instead predict the multiple-scattering field with a neural operator and differentiate through it during inverse rendering, while computing transmittance and single scattering directly. A further challenge arises during optimization: an operator that predicts the field accurately on held-out media can still produce poor derivatives at the density states visited by the optimizer. We address this by fine-tuning on states generated during inversions driven by the operator, using Monte Carlo scattering-field supervision. On held-out optically thick clouds, our method improves relighting over frozen-source tomography at equal time and outperforms differentiable path tracing given substantially more time. Training on inversion trajectories also improves derivative alignment at visited states and reduces optimization drift.
comment: 21 pages, 11 figures, 18 tables
☆ Dataset Distillation via Drifting NeurIPS 2026
The task of dataset distillation concerns the learning of a small set of synthetic samples that can serve as an effective replacement for a large real dataset when used for downstream tasks. While many solutions have been proposed since the problem's inception, most of them primarily seek to optimize the distilled data with respect to a classifier. In this work, we propose a new distillation method that learns a set of samples such that they directly approximate the per-class data distribution. Specifically, we "drift" our distilled data in the latent space of a feature extractor such that they move both toward the real data and away from each other without using a classification objective. We find that our drifted data cover the real distribution significantly better than existing methods while still serving as powerful training samples for classification. Furthermore, we showcase our method's use as a visualization tool for better interpreting models and extend it beyond images to other modalities. Our work advances the frontier of dataset distillation, offering a novel synthesis method and proposing distribution-based metrics as a complementary evaluation. More broadly, our findings help address the growing need to understand both data and models by uncovering more about how they interact.
comment: Accepted at NeurIPS 2026. Project Page: https://drifting-distillation.github.io/
☆ Asymmetric Multimodal Learning with Decomposed JEPAs
Cross-modal alignment is the standard objective for building multimodal representations. We show that alignment as a \emph{sole} objective is lossy wherever modalities are informationally asymmetric---where neither can express what the other carries. Visual question answering is an obvious case: a question is a selective relational operator probing a dense visual field whose fine-grained features no query recovers, so what alignment preserves is not what the task requires. We prove an \textbf{Alignment Redundancy Bound}: alignment-only Dirichlet energy minimization retains at most the redundant information atom $\Red(Z; X_1, X_2)$, annihilating private modality-specific structure $\Unq(Z; X_m)$ and cross-modal synergy $\Syn(Z; X_1, X_2)$.
We introduce \textbf{Decomposed JEPA (D-JEPA)}, which factorizes latents into an aligned shared core and private modality-specific fibers. The shared core retains most of the structure alignment targets; the private fibers recover what alignment discards, via contextual masked prediction and asymmetric cross-subspace independence. Under information asymmetry, alignment-only training undergoes catastrophic semantic collapse on compositional reasoning benchmarks, surviving only on binary queries. D-JEPA averts this collapse, substantially improving downstream accuracy over both alignment-only and alignment-with-reconstruction baselines, and restoring a positive multimodal synergy gap in place of the baseline's inverted gap. Information-theoretic audits confirm that D-JEPA recovers most of the mutual information available from a frozen backbone, and causal fiber-swap interventions demonstrate that the recovered information is causally localized in the private fibers.
comment: 15 pages, 12 tables, 3 figures
☆ When Chekhov's Gun Fires Anyway: Over-Realization in Text-to-Image Models
Jiahao Huang, Kon Woo Kim, Song-Duo Ma, Yuka Kitamura, Felix Wagner, Yunshan Huang, Sunisth Kumar, Xanh Ho, Yun-Ang Wu, Zhijie Chen, Yunjia Xu, Junfeng Jiang, Zefan Yu, Fei Cheng, Akiko Aizawa
In text-to-image generation, a target that the prompt mentions often appears in the generated image in a way the prompt does not call for. We call this failure over-realization. Mental-spaces theory points to the situations where it can happen: under a negation, inside what a character believes or imagines, as the vehicle of a figurative comparison, or behind a viewpoint that hides the target. Prior work has studied only negation in depth, and the conventional alignment and preference metrics do not penalize the failure. We build OverReal to measure it: OverReal-Gen, 698 prompts covering the four situations, on which we measure each generator's over-realization rate; and OverReal-Det, 2,841 generated images human-labeled for whether and how the target is over-realized, on which we verify the automatic judge that produces this rate. Our evaluation of eight text-to-image systems reveals that all of them over-realize, and that conventional metrics such as CLIPScore reward the failure rather than penalize it. Prompt expansion only partly mitigates the failure. Inside the generator, the diffusion backbone attends to a mentioned target almost as much as to a requested one. The cue words that should hold the target back receive less attention than ordinary words.
☆ Safety as a Budget: Calibrating the Control Channel of Constrained GRPO for Vision-Language Models
Reinforcement learning can teach vision-language models (VLMs) to decline harmful image-text requests, but how safe the result becomes is set indirectly, through reward weights found by trial runs. Stating safety as budgets on the rate of harmful assistance and the rate of over-refusal, enforced by Lagrange multipliers inside GRPO with an off-the-shelf scorer as the cost, is the natural fix, and it fails twice. The multiplier compares the scorer's mean with the budget, and scorers built for ranking get the mean wrong: a fixed-point analysis shows that the mean bias, not the budget, then decides whether the multiplier settles, saturates or switches off. And standard GRPO standardizes the penalized reward within each group, which cancels the multiplier before it reaches the gradient. Two-channel constrained GRPO repairs both: the base reward keeps the raw score for the within-group comparison, the penalty and the constraint estimate pass through a monotone map fitted on judge-labeled training responses, and the advantage keeps the multiplier. On a Qwen3-VL-2B policy with a training-free readout of an open 4B model as the scorer, held-out harm on 1,037 adversarial prompts falls from 8.9% under standard GRPO to 4.8%, lower in every seed, where the uncalibrated budget gives 7.6%; over-refusal stays at most 2% on 400 benign queries and MMBench is unchanged. The gain is largest where the intent sits in the image (10.1% -> 5.1%) and grows on the unseen HADES benchmark (28.9% -> 12.8%). Order-preserving distortions of the scores reproduce each predicted controller state in every seed, and the results hold under a map fitted on another judge's labels, an image-reading judge and a human audit.
comment: 10 pages, 5 figures, 5 tables
☆ Improved CNN-Transformer Network for Neutron Image Super-Resolution
The spatial resolution of neutron images is of great importance for material defect diagnosis, structural characterization, and nondestructive testing. However, owing to limitations in neutron beam quality, scintillator conversion efficiency and light scattering, detector pixel size, and noise levels, conventional neutron images often suffer from speckle noise, blurred edges, and missing details, making them inadequate for high-precision analysis. To address these issues, an improved super-resolution model combining convolutional neural networks (CNNs) and Transformers is proposed. First, to overcome the difficulty of obtaining high-quality and clear training images for neutron imaging, a pretrained image restoration model is fine-tuned using real neutron images and then employed to enhance existing neutron images, thereby providing high-quality reference images for subsequent super-resolution model training. Second, an improved CRAFT-based super-resolution network is developed by incorporating a hybrid prior-guided channel fusion module (HPCF), a mean-standard deviation dual-statistics squeeze-and-excitation block (MS-SEBlock), and a partial convolution feed-forward network (PCFN), with the aim of improving the reconstruction of important information, such as textures and fine details, in neutron images. Simulation results demonstrate that, compared with bicubic interpolation, SRCNN, SRGAN-VGG, and other methods, the proposed network achieves superior performance in both full-reference metrics, including PSNR and SSIM, and no-reference metrics, including AGM and SMD. The visual quality and structural fidelity of the reconstructed images are also significantly improved. This study provides a new approach to improving neutron image quality.
comment: 11 pages, 5 figures, 3 tables
☆ Towards Democratising Low-Data Medical Image Classification: CPU-Trainable ZACH-ViT and Its Geometric Variants
Medical image classification in resource-constrained settings must contend with limited annotations and computational capacity. We evaluate four ZACH-ViT variants with linear, Euclidean prototype, Poincare and spherical heads alongside two compact multiple-instance-learning baselines. All models are trained from scratch with fewer than 260,000 parameters across seven MedMNIST+ datasets and three external benchmarks (PCam, ISIC2017 and Shenzhen tuberculosis radiographs). The low-data protocol uses 50 training images per class and five seeds, with validation-selected checkpoints at 23 and 47 epochs; a full-data reference uses one seed. Compact-TransMIL achieves the highest ten-dataset mean test Macro-F1 at 23 epochs (53.34%) and with full data (73.93%). At 47 epochs, ZACH-Linear narrowly leads (57.76% versus 57.54% for Compact-TransMIL). Extending training improves Macro-F1 in 239 of 300 paired comparisons. On ISIC2017, hZACH-Spherical has the highest 47-epoch low-data Macro-F1 (44.09% with SD 4.09%), whereas ZACH-EuclideanProto leads with full data (51.27%). We additionally conduct 40 CPU-only training runs on ISIC2017 and Shenzhen using four CPU threads; model-only single-image inference requires approximately 4-9 ms, while image decoding and preprocessing substantially increase end-to-end latency. The results demonstrate the feasibility of CPU-only training on the tested platform and the task- and regime-dependence of compact medical imaging under strict resource constraints.
comment: 17 pages, 4 figures
☆ ReFineR: Single image to dense point cloud registration at pixel level with point splatting
Cameras and LiDARs (Light Detection and Ranging) provide complementary observations of the same scene: cameras capture rich appearance cues while LiDAR sensors provide accurate geometric measurements. In this work, we address the problem of fine image-to-point cloud registration from a single image and a roughly estimated camera pose. On robotic platforms, residual reprojection errors of several pixels are commonly observed even after calibration due to temporal desynchronization, mechanical vibrations, or inaccuracies in point cloud reconstruction. Such errors can lead to decimeter-level geometric discrepancies at moderate distances and significantly affect multiple downstream tasks, such as dense cross-modal supervision, texture mapping, and semantic label transfer. To address this challenge, we introduce ReFineR: an offline framework to finely register individual images to pre-reconstructed point clouds. ReFineR builds upon differentiable point splatting rasterization and a smooth Gradient Orientation Measure (GOM) to iteratively refine the camera pose through analysis-by-synthesis optimization. We evaluated the method on a curated dataset of manually calibrated images collected from heterogeneous data sources. Furthermore, we show that an automatic point cloud rescale factor can be estimated prior to optimization, enabling robust refinement across a wide range of scenarios using a single scale-invariant formulation. The project page can be found at https://refiner-project.github.io/
☆ Do the Details Help the Whole? One-Way Transfer from Person Actions to Group Intent in VLMs on JRDB-Social
Social scene datasets label people at two levels: what each person is doing and what the group is there for. Training on both levels at once is often assumed to help each of them. We test this assumption for vision-language models (VLMs) on JRDB-Social, treating person actions as the details and group intent as the whole. We fine-tune Qwen3.5 4B under eight training arms with three seeds, repeat the core arms on Qwen3.5 9B and Qwen3-VL 4B, and judge every arm with a verdict rule fixed before testing. Writing each member's actions before the intent raises intent macro-F1 from 30.1 to 31.0, and writing the intent first reaches 31.2, while action F1 stays within 0.2 points of an action-only model. Under the selected recipe, no arm trained with intent improves actions. The gain requires correct, image-grounded actions: empty action lists or actions borrowed from another group with the same intent lower intent F1. Forced-prefix tests, occlusion, member masking, linear probes and component grafting show that the intent follows the written actions, relies more on people and partly resides in the vision tower. Supplying actions as input instead of as targets hurts both levels.
☆ ActWorld-JEPA: Action-Conditioned JEPA World Models Meet Latent Planning for End-to-End Autonomous Driving
Ziying Song, Shengkai Zhang, Caiyan Jia, Lei Yang, Lin Liu, Yuchen Liu, Ziyang Liu, Zirui Li, Gang Wang, Peiliang Wu, Li Wang, Yadan Luo, Chen Lv
Joint-Embedding Predictive Architectures (JEPAs) enable future prediction in latent space without pixel reconstruction. However, latent predictive consistency alone does not establish the relative driving quality of candidate plans. We present ActWorld-JEPA, a joint-embedding formulation for candidate-based driving. Where Plan-Conditioned JEPA (PC-JEPA) uses global and step-wise trajectory intentions to guide recurrent latent prediction, then combines the predicted future with the current scene through gated cross-attention; Relational Residual Reranking ($R^3$) models agreement and discrepancy between plan and future representations under planning supervision, yielding bounded, gated residual corrections to the base planning scores. This formulation connects latent prediction to candidate selection using only observed context and proposed trajectories at inference. Benchmark-specific implementations report 94.1 PDMS on NAVSIM v1 navtest, with the final selector calibrated using aggregate NAVTEST feedback, and 91.6 and 42.7 EPDMS on NAVSIM v2 navtest and navhard, respectively. The nuScenes evaluation reports a mean $L_2$ error of 1.20 m and a predicted collision rate of 0.76% over 1--6 s. This work provides an effective path from JEPA's predictive representations to driving decisions informed by anticipated consequences.
comment: 19 pages, including appendix. Code: https://github.com/adept-thu/ActWorld-JEPA. Project: https://adept-thu.github.io/ActWorld-JEPA/
☆ Pl@ntNet machine-to-machine identification service with Pl@ntNet API: enabling automated biodiversity monitoring through large-scale plant identification
Benjamin Bourel, Jean-Christophe Lombardo, Alexis Joly, Mathias Chouet, Thomas Paillot, Antoine Affouard, Vanessa Hequet, Herv{é} Go{ë}au, Pierre Bonnet
1. Monitoring plant biodiversity requires identifying large numbers of plants across space and time. Although artificial intelligence (AI) can now identify plant species from images with high accuracy, most existing systems rely on users manually submitting images. As ecological monitoring increasingly generates large image collections from cameras, drones, robots and online platforms, there is a growing need for plant identification services that can be directly integrated into automated data-processing workflows. 2. Here, we present Pl@ntNet's machine-to-machine AI plant identification service. This service uses an application programming interface (API) to provide programmatic access to Pl@ntNet's plant recognition models and the associated taxonomic processing. This service enables the implementation of automated identification workflows without human intervention and can thus directly link large image collections and automated observation systems to an AI-based plant identification system. 3. We establish that this service, which covers 85,000 species from around the world, (1) has a species prediction accuracy of 0.89, (2) process images up to 666 times faster than workflows requiring human intervention, (3) and has processed over 84 million identification requests since its launch, including more than 40 million in 2025 alone. We demonstrate how it has already enabled the implementation of a new generation of ecological workflows: identifying plants from images of insects to reconstruct the plant-pollinator interactions, analysing geolocated images from social media to study plant distribution, detecting invasive plants using images taken at the roadside, and identifying weeds using agricultural robots. Finally, we illustrate its application using a reproducible example: identifying plants from insect photos on iNaturalist. In approximately 30 min on a standard laptop, 2,424 images were processed to generate a dataset of 708 plant-pollinator interactions, with a weighted precision of 0.98 for flower detection and an accuracy of 0.88 for genus plant identification. 4. In conclusion, this machine-to-machine species identification is already operating on a considerable scale and meets a genuine need. All of this paves the way for a new paradigm in biodiversity monitoring, in which automated sensors and image collections can be directly connected to AI-based identification services.
☆ Context-Predictive Low-Rank Adaptation for Face Presentation Attack Detection
Aminu Yusuf, Shakeel Muhammad Ibrahim, Sulaimon Oyeniyi Adebayo, Muhammad Sohail Ibrahim, Muhammad Shahid Jabbar, Shujaat Khan
Face presentation attack detection requires representations that distinguish physical spoofing cues across changes in acquisition devices, subjects, and attack media. Large visual encoders offer transferable representations, but conventional binary fine-tuning supervises their final decisions without explicitly constraining how local facial evidence is represented under partial observation. We present CP-JEPA, a context-predictive adaptation framework that combines low-rank tuning of a pretrained CLIP image encoder with a joint-embedding predictive objective. A student predicts complete-image teacher features at masked spatial locations, while unmasked classification, masked classification, and prediction consistency retain the connection to the live/spoof task. An exponential-moving-average teacher supplies evolving latent targets. The auxiliary components are used only during training; inference requires only the LoRA-adapted CLIP visual encoder and binary classifier. {We evaluate CP-JEPA under single-source and multi-source OCIM transfer, additional CelebA-Spoof training, transfer to CelebA-Spoof, two-source training, and unseen-attack evaluation on SiW-Mv2.} On the standard OCIM benchmark, CP-JEPA achieves 3.74\% mean HTER and 98.73\% mean AUC. {Single-source evaluation achieves 11.36\% mean HTER across twelve source--target combinations.} With CelebA-Spoof added to the three OCIM training sources, CP-JEPA reduces mean HTER from 3.45\% to 2.35\% and increases mean AUC from 98.68\% to 99.32\%. On SiW-Mv2, it achieves 3.36\% mean HTER across the held-out attack categories. Controlled loss ablations examine the contributions of latent prediction, masked classification, and prediction consistency. The results show that context-predictive adaptation can improve cross-dataset PAD performance, although the gains vary with the training sources and evaluation setting.
comment: This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible
☆ UniSpark: Exploring Hybrid Attention for Multimodal Understanding and Generation
Jiancheng Pan, Yuzheng Liu, Hengyuan Cao, Qiyuan Zhang, Yangyan Li, Biao Gong, Zheng Zhang, Ka Long Cheng, Xinyang Wang, Yujun Shen
We present UniSpark-9B, a unified multimodal model that investigates hybrid attention for visual understanding and generation. Built on a Mixture-of-Transformer-Experts backbone with approximately 9B parameters in each branch, UniSpark combines full attention for explicit token interactions with GatedDeltaNet linear attention for compact context propagation. Autoregressive text prediction and visual flow matching jointly support visual question answering, text-to-image generation, instruction-guided editing, and interleaved text-image prediction. In interleaved generation, noisy visual targets are isolated from subsequent predictions, while clean visual representations provide context for later outputs. A blockwise protocol enforces this distinction through attention masks in full-attention layers and recurrent-state retention rules in linear-attention layers. The selected training sources contain 159.7M generation samples, 213.4M understanding samples, and 624.8K interleaved sequences. A four-stage curriculum jointly trains multimodal understanding and visual generation, refining answer accuracy and instruction following while extending image synthesis to 2K. UniSpark scores 2470.3 on MME-S and 0.83 on GenEval. Qualitative examples further illustrate its capabilities in interleaved image generation, video synthesis, and video editing. Together, these results demonstrate that hybrid attention provides an effective backbone for unified multimodal understanding and generation.
comment: Technical Report, 31 pages
☆ PiLLVD: A Real-World Low-Light Video Dataset for Dynamic Scenes Captured with Off-the-Shelf Cameras
Low-light image and video enhancement has received increasing attention in recent years. However, most existing enhancement methods are developed and evaluated on still images, while progress in low-light video enhancement remains limited by the lack of real-world paired video datasets. Acquiring paired low-light and normal-light video sequences is challenging because it requires spatially and temporally aligned captures while avoiding motion blur, exposure mismatch, and correspondence errors in dynamic scenes. To address this gap, we introduce PiLLVD, the Pi-based Low-Light Video Dataset, a paired low-light video dataset captured using an accessible Raspberry Pi-based multi-camera system with off-the-shelf cameras. PiLLVD provides paired low-light and aligned normal-light reference sequences for dynamic indoor and outdoor scenes, including diverse human and object motion under varying illumination conditions. Unlike existing datasets that are often image-based, static, synthetic, or limited to simple motion patterns, PiLLVD focuses on motion-rich scenes where frame-to-frame stability and motion-related artefacts are important for evaluation. In addition to presenting the dataset and acquisition pipeline, we benchmark representative pre-trained low-light video enhancement methods on PiLLVD under a zero-shot setting. The results show that existing methods do not generalise uniformly across the proposed sequences, highlighting the challenges posed by the proposed dataset and its value for both benchmarking existing methods and supporting the development of future low-light video enhancement algorithms.
comment: 14 pages, 7 figures, 5 tables. Dataset available on Zenodo: https://doi.org/10.5281/zenodo.23015904
☆ From State of THE ART to Human Prediction Algorithms
Human curiosity and longing for the future have driven human progress from ancient times to the present. With the advancement of AI technology, human prediction algorithms have also drawn increasing attention from researchers. However, as AI technology continues to advance, we observe a counterintuitive phenomenon. Take human trajectory prediction algorithms as an example. Although models' quantitative prediction performance has improved, yielding one state-of-the-art model after another, the consideration of human behaviors, future thoughts, and social interactions themselves presents a gradually diminishing trend in the construction of these prediction algorithms, buried beneath numbers and metrics. In this manuscript, taking interactive art and human trajectory prediction algorithms as examples, we attempt to adopt a different analytical perspective to analyze the structural and developmental commonalities between these two fields. By examining specific interactive artworks and prediction algorithms, we aim to explore how human intuitions and art are associated with AI algorithms through three aspects, including perception, interaction, and timing, thereby providing new insights for the understanding and further research in this domain. We also observe an interesting phenomenon that certain art-inspired prediction algorithms can actually deduce intriguing patterns of human behaviors from the data, offering new potential avenues for understanding the relationship between human activities, state of The Art, and AI algorithms.
☆ Can an LVLM Perceive Event Boundaries Like a Human?
Large Vision-Language Models (LVLMs) reason effectively about video content, but whether they perceive its event boundaries in a manner consistent with humans remains unclear. We first investigate this through Generic Event Boundary Detection (GEBD), which identifies event boundaries naturally perceived by humans in continuous video, without relying on predefined event categories. On Kinetics-GEBD dataset, LVLMs miss many boundaries when predicting directly from frames, but effectively refine candidates proposed by an unsupervised discovery method, achieving state-of-the-art unsupervised GEBD performance. We extend this discovery-refinement approach to hour-long egocentric video, where boundary discovery requires a broader semantic scale and whole-video LVLM refinement becomes impractical. We address these challenges through coarse-to-fine temporal discovery and localized LVLM accept/reject judgments. Human evaluation shows that the LVLM's agreement with group consensus falls within the range observed among human annotators, with disagreements concentrated on low-consensus cases. Using this pipeline, we construct EgoBoundary, a benchmark for event boundary detection in hour-scale egocentric video. Representative supervised and unsupervised methods degrade substantially on EgoBoundary, highlighting the challenge of long-horizon event boundary detection.
comment: 15 pages, 5 figures, 8 tables
☆ HIFC-IQA: Train-Free Cross-Domain Image Quality Assessment via Dual-Process Cognition NeurIPS2026
The gap between controlled synthetic distortions and complex real-world degradations introduces a critical domain shift challenge for No-Reference Image Quality Assessment. Although Unsupervised Domain Adaptation (UDA) approaches attempt to address this issue, they remain constrained by substantial retraining overhead and cumulative error propagation inherent in iterative pseudo-labeling. Inspired by the dual-process theory of human cognition for remarkable generalization ability, we propose Hierarchical Intuitive and Fuzzy Consensus (HIFC), a completely train-free Cross-Domain IQA framework. HIFC employs asymmetric perceptual decomposition to disentangle semantic content from distortion-sensitive representations. The final quality score is fused through the synergy of two complementary branches: a memory consensus branch that retrieves a stable Fuzzy Quality Centroid from a synthetic reference gallery, and an intuitive branch that captures aesthetic polarization via zero-shot vision-language alignment. Extensive experiments show that HIFC achieves state-of-the-art performance across multiple benchmarks, attaining the highest average SRCC and PLCC among competing UDA methods. Most notably, in the challenging synthetic-to-authentic cross-domain setting, HIFC achieves a substantial performance margin over competing methods, demonstrating remarkable robustness and generalization.
comment: Accepted by NeurIPS2026
☆ DesignTaste: Benchmarking AI-Generator Visual Design Against Award-Wining Works
Zhou Chen, JiaXiang Li, DanYing Pan, Ye Li, TianRui Yang, EnBao Zhang, YaNan Xiao, LiXin Sun, Anqi Liu, Dong Zhao, JianYuan Li, YuanYuan Ding, Kai Kong, Xinwen Xu, HuiTang Du, Min Zhou, Zhiqiang Li
Text-to-image benchmarks commonly assess text-image alignment or overall preference. Whether models can produce good designs when a brief leaves the appearance open remains less well understood. We introduce DesignTaste, a visual design benchmark that compares generated images with award-winning works. From these works and their publicly available context, we derive 403 design briefs: 100 poster tasks from 100 Beste Plakate 2025 and 303 tasks spanning 20 categories from ADC 2026. Six leading commercial text-to-image models each generate an image from the same brief. A single model judge scores each output and the corresponding award-winning work independently on five evaluation dimensions. On each dataset, only ArtArk achieves a higher mean overall score than the award-winning works. All six models score lower in Creative Originality on both datasets. These results show that producing finished, visually appealing images does not imply that a model can develop good design solutions.
comment: 31 pages, 12 figures, including appendices. Code and data: https://github.com/ArtArk-AI/DesignTaste
☆ PhysCapBench: A Cross-Domain Benchmark for Physically Grounded Video Captioning
Haofeng Wang, Pengcheng Wen, Ruiqi Li, Yuhao Zhang, Tiange Zhang, Boren Zheng, Xuanyi Liu, Siyu Tao, Danni Yu, Gan Lin, Haocheng Tang, Tianzizhe Guo, Mingyuan Su, Boyuan Chen, Zhongyuan Wang
Physical AI aims to learn transferable physical knowledge from real-world experience to understand, predict, and act in the physical world. Internet videos and embodied demonstrations provide complementary observations of motion, interaction, and state evolution. A foundational data challenge is to measure and preserve this physical information as training supervision. In video-language pretraining, captions express observed processes in language, yet a semantically correct description can omit the interactions and transitions that define those processes. Rich physical experience in video does not automatically translate into sufficient language supervision. This gap calls for a yardstick to assess annotation quality, compare captioning strategies, and guide training-data refinement. Motivated by our data construction for world-model pretraining, we introduce PhysCapBench, to our knowledge the first unified physical-caption benchmark spanning all three domains: third-person scenes, egocentric manipulation, and handheld-gripper demonstrations. It provides human-verified annotations for 1,500 video sequences and evaluates accuracy and coverage through seven physical dimensions. We examine downstream utility through video reconstruction and manipulation-policy fine-tuning. Physical captions reduce reconstruction LPIPS by 8.6-9.5% on third-person videos and 26.9-30.9% on egocentric videos relative to metadata captions. Training-instruction enrichment improves both PhysCapBench scores and policy success, including a 7.5-percentage-point gain for pi0.5 on a RoboTwin 2.0 subtask. PhysCapBench makes the physical content of language supervision inspectable and comparable, providing a cross-domain basis for its evaluation and improvement.
comment: Project page: https://hfwang2001.github.io/physcapbench-site/
☆ I See What You Want: Decoding Hierarchical Operator Intent from Visual Teleoperation History
Vision-language models (VLMs) have become a powerful foundation for robotic scene understanding, task planning, and language-conditioned control. A natural question follows: can these models also decode what a human operator intends while teleoperating a robot, without any explicit instruction? We formulate this as hierarchical intent prediction, jointly inferring the ongoing atomic action and the broader mission it serves from a short visual teleoperation history. To study this, we introduce VI AMO (Vision Intention Action and Mission Ongoing), a dataset of 6,600 episodes from eleven robot sources with 13,653 temporally aligned action annotations and 9,702 mission annotations. Large proprietary models struggle with this problem: on held-out episodes, GPT-4o reaches Sentence-BERT similarities of only 0.563 and 0.470 for action and mission, and GPT-5.6 Luna achieves only 0.581 and 0.554. Fine-tuning a VLM on VI AMO yields our Vision-Intention Model (VIM), which raises these scores to 0.825 and 0.821 and still reaches 0.726 and 0.717 on the 35.4% of test actions with both labels absent from training. Finally, as a downstream application, VIM's predicted mission enables a vision-language-action (VLA) policy to take over from the operator mid-task in a real-world environment unseen during VIM fine-tuning, reducing human teleoperation time by 34.6% relative to GPT-4o and 39.8% relative to GPT-5.6 Luna, while cutting intent-prediction latency by roughly 70% on average.
comment: 8 pages, 6 figures. Project page: https://see-want.github.io/
☆ Model Distribution-Aware Multimodal Dataset Distillation NeurIPS 2026
Multimodal dataset distillation (MDD) aims to reduce the cost of vision-language pre-training by synthesizing compact image-text datasets. Although recent feature-matching methods have improved distillation efficiency, synthetic data may become specialized to the model representations encountered during optimization, limiting generalization to unseen model states. We propose Model Distribution-Aware Multimodal Dataset Distillation (MDA-DD), which leverages supervision from a dynamic staggered queue of models at different training stages. We further introduce asymmetric dual projection for complementary alignment in visual and linguistic spaces and an uncentered second-order moment objective that retains mean-related cross-modal information. Experiments on Flickr30K and MS-COCO demonstrate consistent retrieval improvements over existing methods across synthetic-data budgets. With only 100 synthetic pairs, MDA-DD improves IR@1 and TR@1 on Flickr30K by 12.9 and 18.2 percentage points, respectively, over the strongest baseline at the same budget, outperforming all compared methods using 500 pairs on both datasets across all six retrieval metrics.
comment: NeurIPS 2026
☆ VBot Exo2Ego Policy: Lightweight Data Collection and Exocentric-to-Egocentric Video Generation
We present VBot Exo2Ego Policy, a video generation model that turns third-person videos into first-person observations of a selected actor, offering a practical way to expand first-person operation data from abundant real-world footage. To support this goal, we introduce a lightweight data collection pipeline and mobile app that support first-person recording, third-person recording, and synchronized capture of both views. This setup combines accessible single-view collection with paired RGB and camera geometry for learning cross-view correspondence. Building on a pretrained video diffusion transformer, our model incorporates geometric guidance and auxiliary supervision from semantic masks and hand keypoints to facilitate optimization and improve the fidelity of hand-object interactions. Camera geometry is injected stochastically during training and gradually removed, allowing inference without target-camera calibration. To accommodate the diversity of Internet videos, we combine source-video perception and reasoning, optional actor references, and multi-shot adaptation to identify the intended actor and follow an operation across camera cuts. The collection system and generation model provide a foundation for expanding first-person operation data across tasks, scenes, and participants.
comment: 27 pages, 11 figures
☆ RICapBench: A Fine-Grained Visual Understanding Benchmark Dataset for Vision-Language Models in Remote Sensing
Advances in vision-language models (VLMs) have improved remote sensing image analysis and supported geospatial decision-making. However, fine-grained understanding of remote sensing imagery remains challenging for VLMs. Large variations in object scale, complex geospatial relations among instances, and limited visual detail hinder precise instance localization, visible attribute identification, and spatial relation understanding, potentially leading to descriptions that lack visual grounding. Existing datasets and benchmarks insufficiently assess end-to-end instance localization and captioning without target-specific textual cues. To support this evaluation, we introduce RICapBench, a dataset and benchmark that unifies instance localization and fine-grained captioning in remote sensing. RICapBench contains 245,466 instance-level annotations and 498,005 spatial relation tuples spanning general remote sensing, UAV remote sensing, and dense small-object scenarios. Each annotation pairs an object's oriented bounding box with a natural language description of its visible attributes and spatial relations to surrounding instances and environmental elements. To evaluate the accuracy and completeness of spatial relation captioning, we propose a remote sensing spatial relation parser and the RelCap-F1 metric. We combine RelCap-F1 with instance localization and attribute captioning metrics into a composite metric, RICScore. Our evaluation of 15 advanced open-source and closed-source VLMs identifies instance localization as a major bottleneck for these models. LocateAnything achieves the highest RICScore of 31.68, but scores lower on spatial relation captioning than other task-fine-tuned open-source VLMs. Code and dataset are available at https://github.com/Xldln/RICapBench.
comment: 13 pages, 4 figures, 2 tables
☆ Gravity and Flight: Learning to Navigate Fine-Grained Remote Sensing Scenes
Vision-language models are increasingly capable, yet their reasoning in complex remote sensing scenes can remain coarse-grained and insufficiently grounded in task-relevant visual evidence. We propose Gravity and Flight (G&F), a framework for scene specialization with limited supervision that separates visual and reasoning verification while coupling policy updates. It combines Evidence-of-Thought supervised fine-tuning (EoT-SFT) and Relational Transfer Policy Optimization (RTPO). EoT-SFT teaches fine-grained observation and faithful reasoning through description, analysis, and supported conclusions. RTPO independently samples responses to scene-matched image pairs with fine-grained visual differences under a shared task. Its rubric-based verification combines Gravity, which checks visual fidelity and description coverage, and Flight, which takes descriptions as premises to assess evidence use and relations between paired judgments. Their scores and task correctness yield a shared pair-level advantage that couples policy updates from both responses. Using 1,008 demonstrations and 500 training pairs for harbor specialization, G&F achieves a mean score of 69.12 across six evaluation sets, exceeding the backbone and supervised initialization by 9.76 and 4.56 points. Matched controls show gains over single-image and joint-input RL, and removing either verification view reduces performance. Applying G&F to urban specialization also improves on the base model and supervised initialization across three evaluation subsets. Project details and updates are available at https://github.com/Z0ngL1ng/Gravity-and-Flight.
☆ Teach with Blinks: Breaking Linguistic Momentum in On-Policy Distillation for Fine-Grained Multimodal Perception
On-policy distillation (OPD) has emerged as an effective post-training approach to improving fine-grained perception in multimodal large language models by amortizing costly inference-time zooming operations into a single-pass recipe. However, vanilla OPD is inherently susceptible to linguistic momentum: the teacher can follow the student's erroneous prefixes, thus weakening corrective supervision. To address this, we propose Blink On-Policy Distillation (BlinkOPD), in which the teacher "blinks" at every token: it reads the same prefix once with its eyes open and once with its eyes closed. A first-glance gain further strengthens the correction on rollouts that conflict with the teacher. By extracting more effective supervision from each on-policy trajectory, BlinkOPD achieves state-of-the-art average performance across six fine-grained benchmarks using only 500 training images (12x less data) than existing OPD methods. Beyond fine-grained perception, BlinkOPD also demonstrates strong zero-shot improvements across ten held-out benchmarks spanning general visual perception, visual hallucination, and multimodal reasoning abilities. Code: https://github.com/Supltz/BlinkOPD
☆ FuGIBench: Evaluating Reliable Preparation in Streaming Video
Changheng Cai, Zhichao Huang, Song Ren, Zhichen Zhao, Hongsen Liu, Chaoyang Zheng, Wenxi Wu, Yuan Fang, Wei Wang, Yang Wang, Yuchen Wang
An assistant watching a worker can prepare for an emerging need before the worker is ready to receive help.As the need evolves, the assistant must maintain preparation until a response is due, or withdraw it if the need disappears. Video anticipation and interactive response benchmarks leave this interval between early evidence and response largely unmeasured. We introduce FuGIBench (Future-Grounded Interaction), a streaming-video benchmark in which a model chooses among waiting, reversible preparation, and response given a causal video prefix and a task contract. The benchmark contains 19,893 synthetic tool-assistance videos across ten tool tasks, in handover, ordinary-activity, and cancelled-need tracks. One primary score credits sustained preparation, incorporates response-interval alignment, and penalizes activity when the reference prescribes waiting. On a frozen set of 1,000 episodes, a reference system with a longer preparation lead has poorer response alignment, while similar handover leads accompany different levels of activity during ordinary work in matched scenarios. Together with cancellation analyses, these results show why evaluating preparation requires tracking decisions as needs develop, become actionable, or disappear.
☆ TransXplainRRG: A Radiologist-Evaluated LVLM for Chest Radiograph Report Generation
Recent advancements in Large Vision-Language Models (LVLM) enhance X-ray report generation, but often neglect clinical context, overlook workflow integration, and lack explainability, which hinders clinical adoption due to interpretability and reliability concerns. This paper addresses the dual challenges of performance and expert explainability in automated X-ray report generation for clinical AI applications. Our proposed approach, TransXplainRRG, utilises an off-the-shelf Swin Transformer model along with a transformer-based text encoder that incorporates patient medical history to generate a radiology report. Further, we explore an expert-guided `inside-out' approach and extract only abnormal findings for radiology report refinement. Thus, this study bridges the gap between high-performance automation and interpretability critical for clinical practice by combining state-of-the-art transformer-based vision encoders, text encoders, and LLMs. TransXplainRRG is trained and tested on the large-scale MIMIC-CXR dataset and further evaluated on the unseen IU X-Ray dataset to demonstrate its generalizability and robustness. We employed the GREEN metric as a complementary measure of clinical consistency, achieving scores of 0.327 on MIMIC-CXR and 0.605 on IU X-Ray, both higher than those achieved by MAIRA-2 under the same evaluation setting. It also performed competitively across standard evaluation metrics, frequently matching or surpassing state-of-the-art models. Additionally, we introduce qualitative evaluation metrics developed from radiologists' viewpoints and conduct a preliminary radiologist evaluation on 100 generated reports to assess their clinical relevance. Quantitative and qualitative analyses further demonstrate the potential of our TransXplainRRG framework for reliable and explainable radiology report generation.
comment: 17 pages, 6 figures, 11 tables
☆ RADIANT: Redundancy-Aware Information Allocation for Universal Multi-Modal Fusion
Universal multi-modal fusion often fails not because modalities are insufficiently aligned, but because the fusion process allocates information poorly: redundant evidence is repeatedly amplified, modality-unique evidence is over-smoothed during alignment, and task-dependent synergistic evidence is insufficiently activated. We present RADIANT (Redundancy-Aware Dynamic Information Allocation Network), a task-general framework that organizes fusion around three complementary roles: suppressing redundant cross-modal overlap, preserving modality-unique evidence, and composing task-conditioned synergy through cross-modal interaction. Specifically, stable matrix-exponential projections are used to reduce interference among shared and modality-unique streams, while task memory with hard positive updates selects which synergistic interactions should be activated for the current fusion regime. Extensive experiments across six heterogeneous fusion tasks show consistent improvements over the compared methods. On FLIR, RADIANT improves downstream detection mAP@0.5 from 91.2% to 96.8%, a gain of 5.6 percentage points (6.1% relative) over GIF. RADIANT also exhibits strong direct cross-dataset transfer, with further gains from inference-time task-memory adaptation while keeping all network weights frozen. Additional ablations and sensitivity studies support the robustness of the allocation mechanism.
comment: 9 pages, 4 figures. Published in ACM Multimedia 2026 (MM '26). Code: https://github.com/StevenWSH27/RADIANT
☆ Description-Guided CLIP Adaptation with Local Evidence Calibration for Few-Shot Plant Disease Recognition
Accurate plant disease recognition is essential for intelligent agriculture, but few-shot settings provide limited labeled samples that may not capture diagnostic symptoms and intra-class variations. Pretrained vision-language models (VLMs), such as CLIP, offer transferable visual-semantic knowledge. However, existing CLIP-based methods often rely on coarse class semantics and global visual representations, overlooking disease-specific diagnostic information and fine-grained local cues. Excessive few-shot adaptation may also cause over-specialization and weaken the generalization ability of large-scale pre-training. We propose a description-guided, two-stage CLIP adaptation method to address these issues. Symptom and comparative descriptions enrich class semantics with typical visual symptoms and discriminative cues between similar diseases. The enhanced semantics guide the aggregation of class-relevant patch evidence to calibrate global image-text matching with fine-grained cues. For adaptation, we first update the vision-language encoders with lightweight parameters, then freeze them and train a linear classifier. At inference, the classifier is used for in-domain prediction, while class prototypes from the adapted text encoder support cross-domain prediction and retain vision-language transferability. Experiments with two backbones show that our method achieves the highest mean accuracy in most few-shot settings on PlantVillage, Cassava, and Pear, reaching 93.16%, 79.30%, and 96.56%, respectively, with ViT-L/14 under 16-shot. On class-aligned cross-domain evaluation across 21 shared PlantDoc classes, it achieves 56.96% accuracy, outperforming all compared methods. These results demonstrate its effectiveness in fine-grained few-shot plant disease recognition and cross-domain generalization.
☆ ReliaGuard: Reconstruction-Aware Watermarking for Source Attribution and Blind Edit Localization
Generative image editing facilitates content creation but complicates source attribution and edit verification. Existing watermarking methods combine copyright protection with tamper localization, yet background reconstruction can weaken spatial evidence even when source identifiers remain recoverable. We propose ReliaGuard, a reconstruction-aware watermarking framework for source attribution and blind edit localization. The writer optimizes keyed feature responses in both the protected image and its variational autoencoder reconstruction under image-distortion and source-preservation constraints. Using only the received image and a secret key, the reader combines watermark responses with local image context through residual evidence refinement and predicts an edit mask through native-resolution-guided multiscale reading. It is trained on complete editing outputs, locally spliced counterparts, and processed versions of both. On an independent test set of 300 source images and 900 editing cases from three image domains, our method achieves edit-region intersection-over-union scores of 85.61% for complete outputs and 94.90% for locally spliced outputs at a threshold calibrated to a nominal 1% background false-positive rate on complete outputs. The complete-output result exceeds the strongest baseline by 45.02 percentage points, while decoded identifier-bit accuracy reaches 100% in both conditions. Across seven families of image processing, the corresponding localization scores are 81.01% and 86.49%, with identifier-bit accuracies of 92.76% and 92.79%, respectively. These results support source recovery and blind edit localization under background reconstruction and subsequent image processing.
☆ Hand in Silhouette: Synthetic-to-Real Multi-View Hand Reconstruction via Binary Masks
Calibrated multi-view hand reconstruction has advanced rapidly, yet the effectiveness of existing methods depends strongly on real captured data, which is costly to acquire, limited in hand-geometry diversity, and often affected by imperfect annotations. Synthetic data offers a scalable alternative with accurate supervision, but a substantial synthetic-to-real appearance gap hinders its use. To address this, we introduce Hand in Silhouette (SilHand), a calibrated multi-view hand reconstruction framework that operates on binary hand masks, obtainable at low cost either automatically or manually, to reduce the synthetic-to-real observation gap. Rather than relying on RGB features for 3D lifting, SilHand projects metric 3D hypotheses into each view to query and fuse mask-ray evidence for hand reconstruction. This mask-centric formulation enables training on a large-scale synthetic dataset comprising 46.8k grasping poses and 60k free-hand samples, providing broad coverage of hand geometry while directly transferring to real captured hand geometry. Across five real-world benchmarks, SilHand improves relative reconstruction accuracy from 49.2% to 63.4% using masks rendered from reference hands. Moreover, reconstruction accuracy improves consistently as the synthetic training set grows, indicating that broader geometric coverage translates into better real-world performance. While achieving such diversity is difficult with real captured data, it becomes practical under our mask-centric formulation. Finally, SilHand remains robust under mask corruption, with reconstruction error increasing by at most 1.63 mm. Together, these results support binary masks as an effective interface for learning hand reconstruction from synthetic geometry.
comment: Project page: https://silhand26.github.io
☆ Selective Generative Restoration for Video Deraining via Uncertainty-Aware Diffusion
Video deraining is challenging due to rain streaks, occlusions, and atmospheric effects. Existing methods mainly follow a deterministic restoration paradigm, assuming that clean scenes can be uniquely recovered from rainy observations. However, rain degradation is spatially heterogeneous. Reliable regions require faithful reconstruction, whereas severely occluded regions contain ambiguous content that require stronger prior-driven inference. Therefore, conventional deterministic restoration often produces over-smoothed results, while global generative restoration may introduce unnecessary hallucinations. We introduce a Selective Generative Restoration paradigm for video deraining, where generative priors are adaptively activated according to restoration uncertainty. Based on this principle, we propose Selective Generative Video Deraining (SGVD), an uncertainty-aware diffusion framework that enables adaptive fidelity preservation and generative recovery. Specifically, SGVD exploits stochastic diversity between dual one-step diffusion predictions to estimate region-level restoration uncertainty, which determines where observation fidelity should be preserved and where generative inference should be activated. Furthermore, an anchor-guided trajectory consistency mechanism is proposed to regularize stochastic diffusion evolution across frames, reducing temporal generation drift and improving video coherence. To achieve practical efficiency, SGVD adopts one-step diffusion inference while maintaining uncertainty-aware generative capability. Extensive experiments demonstrate that SGVD achieves superior restoration performance, perceptual quality, and temporal consistency compared with state-of-the-art video deraining methods.
☆ DEDCA: Test-Time Adaptation for Generalized AI-Generated Image Detection
AI-generated image detectors face significant challenges when deployed in real-world environments, particularly when test samples are produced by unseen generators that deviate from the source training distribution. Existing detectors often rely on a single type of evidence, such as semantic representations or low-level generation traces, making them vulnerable when the corresponding cue becomes unreliable. Moreover, adapting detectors to newly emerging generators usually requires target labels or retraining, which is costly and impractical. To address these challenges, we propose Dual-Evidence Disagreement-Constrained Adaptation (DEDCA), a test-time adaptation framework for generalized AI-generated image detection. DEDCA constructs a dual-evidence detector by combining CLIP-based semantic evidence with trace evidence extracted from shuffled entropy and residual statistics, and introduces a reliability-aware gate to perform sample-specific adaptive fusion. Our key idea is to exploit the disagreement between the two evidence streams as an unlabeled adaptation signal, enabling the detector to adjust to target-domain shifts without relying on target labels or blindly trusting a single prediction stream. Experiments show that DEDCA achieves superior performance on the GenImage benchmark by outperforming state-of-the-art AI-generated image detectors and test-time adaptation methods with 92.0\% average accuracy.
☆ FlowIR: Depth-driven Rectified Flow for Resolving Thermal Ambiguity in Infrared Super-Resolution ICME 2026
Distinguishing different heat sources in low-resolution infrared (IR) images is inherently difficult, as multiple thermal emitters often collapse into ambiguous and overlapping responses. Existing super-resolution methods hallucinate structures from such ambiguous 2D evidence, leading to thermal drift, boundary merging, and unreliable restoration under infrared degradations. We propose FlowIR, a depth-driven rectified flow framework that resolves thermal ambiguity by injecting 3D geometric cues into the generation process. Pseudo-3D depth lifting and depth-guided 3D convolutions disentangle overlapping foreground objects, while cluster-wise inter-attention suppresses cross-source feature contamination. A Thermal-Geometry Consistency Loss further enforces geometric alignment during refinement. Built upon a lightweight rectified flow backbone, FlowIR restores IR images in only a few steps, producing sharply separated, degradation-robust, and geometrically faithful high-resolution results.
comment: Accepted to the 2026 IEEE International Conference on Multimedia & Expo (ICME 2026)
☆ Reconstruct What You Can: Distractor-free 4D Gaussian Splatting WACV 2027
Dynamic reconstruction from casual video must preserve valid subject motion while suppressing transient occluders, shadows, and volumetric effects. Existing methods either suppress valid motion through static-scene assumptions or model all observations jointly, producing ghosting and unstable geometry. We introduce ReCan-GS, a 4D Gaussian Splatting framework that separates coherent scene content from inconsistent observations. It derives a Jacobian-based deformation-sensitivity score from the learned deformation field, using it as an empirical proxy for reconstruction ambiguity. The score guides separation into dynamic Core Gaussians and per-view Distractor Gaussians, or post-hoc pruning of sensitive Core Gaussians. The same criterion supports removal of spatial occluders, photometric artifacts, and evolving smoke or fog without masks, pretrained segmenters, or medium-specific supervision. Experiments on synthetic and real dynamic scenes and synthetic smoke/fog scenes demonstrate its usefulness across the evaluated distractor regimes.
comment: Accepted at WACV 2027
☆ ATune: Unsupervised Photo Retouching Through Image-Editor Controls
Photo retouching models are trained on pairs of original and retouched photographs, usually edited by experts. Expert edits are costly, so the two main public sets, FiveK and PPR10K, hold only 16,161 originals. Good photographs on their own are far easier to find: commercial stock collections and free ones such as Pixabay and Unsplash hold millions. We learn retouching from them alone: we damage each photograph the way a camera or a careless edit would, which gives a worse and a better version of the same photograph. From these pairs, our model, ATune, learns how 13 measured attributes, such as brightness, contrast and sharpness, should change to raise a photograph's quality. The edit is declarative: ATune sets target values for the attributes, and a simple search adjusts the sliders of an out-of-the-box photo editor until the photograph reaches them. Every edit is a list of slider values a person can read and adjust. The same model can drive any editor that makes those changes; we test RawTherapee and darktable. We train on 10M photographs from a commercial stock collection and test on FiveK, PPR10K and SPAQ. Nine quality judges score every edit, and in a user study 157 human raters made 9,253 choices between our edit and another method's edit of the same photograph. When Claude Opus 5 and GPT-6 Astra edit the same photographs in the same editor, raters prefer our edits to GPT-6 Astra's (54.3%) and rate them even with Claude's (50.5%). We also compare with supervised retouchers on their home ground: NeurOP on FiveK and AdaInt on PPR10K, each trained on that set's own expert edits. Raters chose our edits in 46.3% of those comparisons. ATune, a 54M-parameter model trained without a single expert edit, now matches frontier models at setting an editor's sliders.
comment: 27 pages, 12 figures, 18 tables. Main paper 8 pages plus supplementary material
☆ Disentangled Embedding Optimization for Diffusion-Based Dataset Distillation ACM MM 2026
Dataset distillation aims to compress large-scale datasets into compact synthetic subsets that enable efficient model training with comparable performance to the original data. Recent advances show that diffusion models are highly effective for this task due to their strong generative capability. Existing diffusion-based approaches typically perform class-wise clustering and use the obtained cluster prototypes as diffusion inputs to generate distilled samples. However, the uniformly distributed original samples in a category inevitably lead to entanglement near cluster boundaries, pulling the cluster prototypes closer and reducing the inter-cluster separation. Consequently, the diffusion model receives overly similar latent inputs, resulting in synthetic data with limited diversity. To address this issue, we propose Disentangled Embedding Optimization (DEO), a simple yet effective framework that replaces static clustering with trainable embeddings. These embeddings are optimized to match the Gaussian statistics (i.e., mean and covariance) of the original data distribution, while a diversity regularizer is designed to encourage separation among these embeddings to improve the intra-class variability. Extensive experiments demonstrate that our proposed DEO can generate more diversified and representative distilled data than existing cluster-based diffusion methods, achieving state-of-the-art performance and superior generalization across multiple datasets.
comment: 10 pages, 7 figures. Accepted at the 34th ACM International Conference on Multimedia (ACM MM 2026)
☆ WorldDraft: If Code Builds the World, Can Video Models Just Render It?
Zhiyu Hou, Wen-Jie Shu, Harold Haodong Chen, Weilin Ruan, Chenfei Liao, Liuzhou Zhang, Disen Lan, Xiaogang Xu, Qi Chen, Ser-Nam Lim
Multimodal coding agents decouple 4D world construction from visual generation, repurposing video models from world generators toward generative renderers. This division of labor raises a fundamental question: if code builds the world, can video models just render it? A reliable generative renderer hinges on two complementary criteria: (i) world preservation, retaining prescribed 4D kinematics and camera paths; and (ii) visual realization, persistently binding text-specified target attributes over time. To evaluate this capability, we introduce WorldDraft, a benchmark of $432$ strictly paired test cases across $48$ executable 4D world families spanning six dynamic challenge categories. Under a $3 \times 3$ factorial design, WorldDraft orthogonally crosses draft granularity (abstract proxy to previs) and appearance alignment (neutral to conflicting) under identical dynamics. Extensive evaluations across $12$ representative models reveal that: (I) world preservation and visual realization remain in stark operational tension; (II) richer draft scaffolding anchors structure but paradoxically impairs appearance fidelity; and (III) conflicting source cues trigger severe source leakage (up to $64\%$), exposing fragile cross-modal disentanglement. To mitigate these bottlenecks, we propose WorldDraft-Primer, a plug-and-play visual prompting strategy that consistently improves both objectives across frozen video renderers. Together, WorldDraft and WorldDraft-Primer lay a foundation for bridging programmatic 4D simulation with video generation, advancing physically grounded generative rendering.
comment: Project: https://haroldchen19.github.io/WorldDraft-site/
☆ SPEM: Scan-Prior Entropy Modeling for LiDAR Point Cloud Compression
LiDAR point clouds exhibit scan-induced spatial regularities that are informative for geometry compression. Existing scan-aware approaches often exploit these regularities by transforming coded geometry into spherical or beam-aligned coordinate systems. However, this couples acquisition priors with geometry representation, altering spatial partitioning and quantization, and potentially preventing exact recovery of the original Cartesian geometry. We argue that acquisition priors should guide probability modeling rather than dictate the coding representation. We propose SPEM, a Scan-Prior Entropy Modeling framework that exploits LiDAR scanning regularities within the entropy model without requiring coordinate transformation. We introduce a multi-frequency spherical rotary positional encoding that captures scan-aligned spatial relationships in both backbone self-attention and decoded-context retrieval. Building upon this encoding, we develop a decoded-memory autoregressive predictor that efficiently exploits previously decoded occupancy information through reusable key/value caches. Experiments demonstrate state-of-the-art performance in both lossy and lossless LiDAR geometry compression, highlighting the effectiveness of scan-prior modeling across different coordinate representations.
☆ GOOSELoc: A Multimodal Benchmark for Ground-to-Aerial Localization in Unstructured Off-Road Terrain
Fabian Deuser, Peter Mortimer, Robin-Nico Kampa, Anna Bößendörfer, Mirko Mählisch, Martin Werner, Norbert Oswald
Cross-view pose estimation estimates a ground camera's geographic pose by matching onboard sensor data with geo-referenced overhead imagery. This is particularly relevant when GNSS is unreliable, for example under forest canopy or in urban canyons. Existing metric cross-view benchmarks, such as VIGOR and cross-view KITTI, are dominated by urban and road scenes with strong human-made structure without off-road terrain. We introduce GOOSELoc, a multimodal benchmark for ground-to-aerial localization in unstructured environments, including forests, fields, and rural tracks. GOOSELoc comprises 9,356 scenes recorded across all four seasons around Munich, Germany, pairing geo-referenced aerial orthophotos with surround-view RGB imagery and 3D LiDAR point clouds. Paved/unpaved labels enable analyzing the impact of terrain structure, while the seasonal recordings allow studying the appearance gap to mostly summer-captured orthophotos. To assess the transferability of existing methods, we adapt and fine-tune four state-of-the-art cross-view localization baselines on GOOSELoc. Despite task-specific fine-tuning, the performance degrades substantially in off-road environments. We therefore propose a multimodal baseline combining surround-view imagery and LiDAR geometry, improving localization performance and providing a stronger reference for future work. Data and baselines are released at https://gooseloc.github.io.
☆ Split-and-scale Latent 3D Representations NeurIPS 2026
Latent 3D representations have significantly expanded the capabilities of modern shape modeling, enabling compact encoding of 3D scenes and supporting powerful 3D generative models. Despite this progress, current representations struggle to capture the fine details of complex objects, fundamentally limited by the capacity of their latents. However, scaling latent capacity at training time is computationally expensive and often infeasible under memory limits, and naively increasing the number of latents at inference fails to make effective use of the added capacity. In this paper, we address both challenges with \textit{split-and-scale}, a method that effectively expands latent capacity at inference while keeping training cost tractable. Our key observation is that by representing 3D scenes as continuous functions of 3D coordinates, a single pretrained tokenizer can be applied at inference to arbitrarily small sub-regions of a scene, allocating the same latent capacity to a smaller spatial extent and thus capturing far higher detail. We validate this insight by conducting extensive ablations to study the trade-offs between compute, memory, latent capacity, and reconstruction quality, demonstrating our inference-time scaling method achieves quality competitive with much larger models that exceed single-GPU memory limits. We further demonstrate that split-and-scale supports high-quality generative modeling, training an autoregressive 3D generator that performs competitively with state-of-the-art baselines.
comment: Accepted to NeurIPS 2026
♻ ☆ Understanding and Learning Prediction Targets in Diffusion Models
Recent advances in diffusion and flow matching models have highlighted a shift in the preferred prediction target -- moving from noise ($\varepsilon$) and velocity (v) to direct data (x) prediction -- particularly in high-dimensional settings. However, a formal explanation of why the optimal target depends on the specific properties of the data remains elusive. In this work, we provide a theoretical framework based on a generalized prediction formulation $u=kx-(1-k)n$, where $x$ represents clean data and $n$ denotes noise. This formulation accommodates arbitrary targets, of which $\varepsilon$-, v-, and x-prediction are special cases. We derive the analytical relationship between data's geometry and the optimal prediction target, offering a rigorous justification for why x-prediction becomes superior when the ambient dimension significantly exceeds the data's intrinsic dimension. Furthermore, our controlled analysis reveals how time conditioning and loss variations systematically modulate this optimal objective. While our theory identifies dimensionality as the governing factor for the optimal prediction target, the intrinsic dimension of manifold-bound data is typically intractable to estimate in practice. To bridge this gap, we propose k-Diff, a framework that employs a data-driven approach to learn the optimal prediction parameter k directly from data, bypassing the need for explicit dimension estimation. Extensive experiments in both latent-space and pixel-space image generation demonstrate that k-Diff consistently outperforms fixed-target baselines across varying architectures and data scales, providing a principled and automated approach to enhancing generative performance.
comment: 26 pages, 5 figures
♻ ☆ Dual Flow: Joint Image-Measurement Interpolation for Inverse Problems SP
Generative model-based approaches to linear inverse problems typically use the acquired noisy measurement to condition or guide sampling in the image domain. In this work, we introduce Dual Flow, a novel joint-flow model for solving linear inverse problems by jointly estimating the clean image and corresponding noise-free measurement from the observation. Specifically, Dual Flow formulates image reconstruction and measurement denoising as a coupled estimation problem, using coupled flows in the image and measurement domains. The acquired noisy measurement is used only to initialize the measurement state; thereafter, image and measurement states are jointly updated based on their current values. By construction, the final image and measurement estimates are guaranteed to satisfy the underlying forward model. We derive the sampling updates for the joint-flow which involves the conditional mean of the clean image given the joint state. We further derive a closed-form expression for a sufficient statistic of the joint state and show that the required conditional mean can be obtained by solving an image-domain Gaussian denoising problem involving operator-dependent, correlated noise. We train a velocity-prediction network using this formulation to estimate the conditional mean from the derived sufficient statistic. Extensive experiments across various linear inverse problems and image datasets demonstrate state-of-the-art performance in both distortion metrics and perceptual quality. Dual Flow requires no hyperparameter tuning beyond the number of inference steps and achieves shorter reconstruction times than competing methods. Additional analyses assess robustness to model mismatch, as well as the ability to generate diverse plausible reconstructions and characterize reconstruction uncertainty.
comment: 20 pages, 9 figures. Our code is available at https://github.com/METU-SPACE-Lab/Dual-Flow
♻ ☆ MedGS: Gaussian Splatting for Multi-Modal 3D Medical Imaging
Kacper Marzol, Ignacy Kolton, Weronika Smolak-Dyżewska, Joanna Kaleta, Żaneta Świderska-Chadaj, Marcin Mazur, Mirosław Dziekiewicz, Tomasz Markiewicz, Przemysław Spurek
Multi-modal three-dimensional (3D) medical imaging data, derived from ultrasound, magnetic resonance imaging (MRI), and computed tomography (CT), provide a widely adopted approach for non-invasive anatomical visualization. However, accurate modeling depends on surface reconstruction and frame-to-frame interpolation, where traditional methods often struggle with image noise and incomplete information between sparse frames. To address these challenges, we present \our{}, a novel framework based on Gaussian Splatting (GS) designed for high-fidelity 3D anatomical reconstruction. Uniquely, \our{} employs a multi-task architecture that simultaneously performs frame interpolation and segmentation using a unified geometric representation. By coupling these tasks, the model leverages dense signals from image synthesis to regularize the geometry, enabling high-quality surface extraction even from a limited number of input frames. Specifically, medical data are modeled as Folded-Gaussians with dual color attributes, supported by an In-Between Frame Regularization (IBFR) mechanism. Experimental results demonstrate that \our{} achieves higher metric scores than implicit neural representations and improves interpolation quality. Repository: https://github.com/gmum/MedGS
♻ ☆ GRCF: Group-wise Ranking and Calibration Framework for Multimodal Sentiment Analysis
Most multimodal sentiment analysis algorithms focus on point-wise regression. While straightforward, it uses absolute values to represent subjective sentiment and overlooks the ordinal nature of sentiment, resulting in unstable predictions and poor correlation. Inspired by the way humans perceive sentiment, pair-wise ordinal learning frameworks emerged to address this gap, capturing relative order by learning from comparisons. However, they have two limitations: First, they assign uniform importance to all pairs, failing to adaptively focus on hard-to-rank samples. Second, they employ static ranking margins, which fail to reflect the varying semantic distances between sentiment groups. To address this, we propose a Group-wise Ranking and Calibration Framework (GRCF) that introduces the Adaptive Error-weighting and Dynamic Margin mechanism, which resolves these limitations by simultaneously preserving the relative ordinal structure and ensuring absolute score calibration. Specifically, Stage 1 introduces a Group-aware Ranking Loss to build a fine-grained ordinal structure by assigning higher optimization priority to confusable pairs. Stage 2 then employs an MAE-driven objective to align the prediction magnitudes. To validate its generalizability, we extend GRCF to classification tasks, including Multimodal Humor and Sarcasm Detection. It achieves state-of-the-art performance on regression datasets while also showing strong generalizability in classification tasks.
♻ ☆ VMonarch: Efficient Video Diffusion Transformers with Structured Attention
The quadratic complexity of the attention mechanism severely limits the context scalability of Video Diffusion Transformers (DiTs). We find that the highly sparse spatio-temporal attention patterns exhibited in Video DiTs can be naturally represented by the Monarch matrix. It is a class of structured matrices with flexible sparsity, enabling sub-quadratic attention via an alternating minimization algorithm. Accordingly, we propose VMonarch, a novel attention mechanism for Video DiTs that enables efficient computation over the dynamic sparse patterns with structured Monarch matrices. First, we adapt spatio-temporal Monarch factorization to explicitly capture the intra-frame and inter-frame correlations of the video data. Second, we introduce a recomputation strategy to mitigate artifacts arising from instabilities during alternating minimization of Monarch matrices. Third, we propose a novel online entropy algorithm fused into FlashAttention, enabling fast Monarch matrix updates for long sequences. Extensive experiments demonstrate that VMonarch achieves comparable or superior generation quality to full attention on VBench after minimal tuning. It overcomes the attention bottleneck in Video DiTs, reduces attention FLOPs by a factor of 17.5, and achieves a speedup of over 5x in attention computation for long videos, surpassing state-of-the-art sparse attention methods at 90% sparsity. Code is available at https://github.com/EricLina/VMonarch.
♻ ☆ Meituan-Robotics-0: A Vision-Language-Action Foundation Model for Desktop Manipulation
Meituan Robotics Team, Caicheng Wang, Chen Zhao, Chengpeng Jin, Duo Xu, Fengjiao Chen, Guanglu Wan, Heqing Yang, Jia Wang, Jiacong Zhou, Jie Hu, Jiyuan Ru, Lingfeng Zhang, Linjiang Huang, Manyuan Zhang, Ningchu Liu, Shijie Zhou, Si Liu, Shuo Li, Tianyi Zhang, Xi Zhang, Xin Chen, Xuezhi Cao, Xunliang Cai, Yanbin Zeng, Yang Yang, Yang Yi, Yanli Tan, Yifu Xu, Yimeng Jia, Yuchen Tang, Yuchen Xie, Yuxiang Zheng, Zerwei Chng, Zilong Huang, Zipeng Ye, Ziyuan Qin, Ziqin Wang
We present Meituan-Robotics-0 (MR0), a vision-language-action foundation model and embodied-agent system for stationary-base desktop manipulation. MR0 combines 147 million vision-language samples with 15,060 hours of curated robot trajectories spanning 12 embodiments. A three-stage training recipe aligns multimodal representations with robot behavior through discrete action-token pre-training, transfers these priors to continuous cross-embodiment control through flow-matching mid-training, and adapts the policy through real-world post-training. We introduce Rollback, which pairs replaced policy segments with human corrections to provide positive and negative supervision for CFG-DAgger without value learning or advantage labeling. A unified end-effector interface supports heterogeneous embodiments, while asynchronous action-chunk merging maintains smooth execution under variable inference latency. An event-driven agent maintains persistent task events and reassesses plans when progress or feasibility changes, enabling cross-task retrieval and conditional task resumption. MR0 achieves 93.6% success on RoboTwin 2.0 and ranks fifth in the RoboDojo VLA track. Under matched HG-DAgger post-training, MR0 achieves higher average success rate and progress score than $π_{0.5}$ across three real-robot tasks. On Ethernet insertion, CFG-DAgger trained with Rollback data reaches 70% success and a progress score of 0.90. Qualitative agent deployments demonstrate cross-task event retrieval and recovery from temporarily unsatisfied task conditions.
comment: 25 pages. Project page: https://meituan-robotics.github.io/Meituan-Robotics-0/ ; Code: https://github.com/Meituan-Robotics/Meituan-Robotics-0
♻ ☆ DiffCVE: Diffusion-based Compressed Video Enhancement
Perceptual quality enhancement of severely compressed videos remains challenging due to complex artifact patterns and substantial information loss. Recent diffusion models have demonstrated strong generative capability for visual restoration, but directly applying them to compressed video often ignores compression degradation characteristics and may introduce structure-inconsistent hallucinations. To address this issue, this paper presents a diffusion-based compressed video enhancement method, named DiffCVE. Coding Prior-enhanced Dual Conditioning (CPDC) branches are designed to jointly model compressed video and coding prior conditions, where coding priors including residuals and motion vectors provide complementary structural and motion guidance during the diffusion denoising process. To make the diffusion process aware of compression severity, a Compression Degradation Semantic Prompting (CDSP) mechanism is introduced to leverage QP-conditioned textual prompts together with LoRA fine-tuning. In addition, a Coding Prior-guided Weighted Fusion (CPWF) module is incorporated into the VAE decoder to fuse VAE encoder and coding prior encoder features with QP-predicted weights. Extensive experiments demonstrate the effectiveness of the proposed method in improving perceptual quality, especially under severe compression settings. The project page and code are available at https://wqmaker.github.io/projects/DiffCVE/.
♻ ☆ ChessQueries: Toward Better Chess Board Recognition
Chess board recognition is the task of mapping the image of a chess board to the information of which piece is on which square. So far this task has two established benchmarks: ChessCog is synthetic, and ChessReD comes from smartphone pictures of a single chess board setup. We introduce ChessQueries, a new method combining a ViT encoder with a DETR-style decoder, which outperforms existing methods. On the ChessReD benchmark, we improve the state of the art from 15.3% to 99.2%, and demonstrate strong capabilities on out-of-distribution datasets. Our method saturates the task on the two datasets, with an average 0.01 wrong squares per board (vs. SotA: 3.4 / 0.15 respectively). We also share a new, harder public dataset, parsed from broadcasted top-level chess tournaments. Code, model weights and the SLCC data have been released.
♻ ☆ TinyAstral: Transferring Self-Evolving Harnesses to Tiny Generalist Manipulation Agents
Kehong Gong, Xiaoxu Zheng, Guanli Hou, Seunghun Lee, Mengqi Guo, Runjia Tan, Zhengyu Wen, Qi Wang, Xiaoyu He, Mingyuan Zhang, Michael Bi Mi
Robot agents built on vision-language models depend on both the planner and its harness: the tools, feedback, memory, and recovery logic around it. Can improvements to a large teacher's harness be carried into a compact agent with lower inference compute cost? We present TinyAstral, a two-stage approach. We first freeze an open-weight teacher and improve its harness through recursive self-improvement (RSI): frontier LLM agents read the teacher's trajectories and propose harness edits, and each edit is accepted only if it passes a paired statistical test. We then freeze the final harness, collect teacher trajectories with reasoning and failure recovery, and distill a small student that replaces the teacher in the same harness. On RoboTwin, harness evolution raises the teacher's success from 33.4% to 63.6%; the final 4B student reaches 62.8%. Across four harness versions, distilled 4B students score close to their corresponding teachers, and later versions yield stronger students, showing that harness evolution gains carry into compact models. The harness gains also transfer to other frozen planners without retraining. On identical episodes, the final 4B student outperforms open 4B agents, the open 27B harness-RSI model SenseNova-RoboRSI, and our own teacher model placed in three competitors' harnesses. It stays close to its teacher on all three training benchmarks, and on three benchmarks never used for RSI or training it scores close to a 552B planner running the same harness. On the measured deployment episodes, it uses about one fifth of its teacher's accelerator time. Project page: https://garfield-kh.github.io/TinyAstral/
comment: 16 pages, 5 figures. Project page: https://garfield-kh.github.io/TinyAstral/
♻ ☆ Disambiguating 2D-3D Correspondences in Gaussian Splatting-based Feature Fields for Visual Localization NeurIPS 2026
While Gaussian Splatting-based Feature Fields (GSFFs) have shown promise for visual localization, this paper highlights that photometrically optimized GSFFs are inherently not suitable for 2D-3D matching. The volumetric extent of each Gaussian induces many-to-one pixel-to-point mappings that destabilize PnP-based pose estimation, while photometric optimization gives rise to superfluous Gaussians devoid of multi-view consistency. To address these issues, we propose SplitGS-Loc, a GSFFs construction framework for localization, which disambiguates 2D-3D correspondences by exploiting Gaussian attributes. Our key design, Mixture-of-Gaussians-based splitting, decomposes each Gaussian into smaller Gaussians, replacing ambiguous many-to-one with precise one-to-one correspondences. In parallel, we exploit composition weights from GS rasterization to select Gaussians that significantly and consistently contribute across multiple views and aggregate discriminative features through strong pixel-Gaussian associations. The resulting compact yet discriminative feature fields enable stable PnP convergence, achieving competitive performance on localization benchmarks. Extensive experiments validate that SplitGS-Loc extends the utility of photometric GSFFs to accurate and efficient localization by exploiting Gaussian attributes, without per-scene training for feature fields or iterative pose refinement. The code is available at: https://github.com/misoleekr/SplitGS-Loc.
comment: Accepted to NeurIPS 2026. Project Page: https://misoleekr.github.io/SplitGS-Loc/
♻ ☆ Efficient Multi-Organ Segmentation Using SpatialConfiguration-Net with Low GPU Memory Requirements
Even though many semantic segmentation methods exist that are able to perform well on many medical datasets, they are often not designed for direct use in clinical practice. The two main concerns are generalization to unseen data with a different visual appearance, e.g., images acquired using a different scanner, and efficiency in terms of computation time and required Graphics Processing Unit (GPU) memory. In this work, we employ a multi-organ segmentation model based on the SpatialConfiguration-Net (SCN), which integrates prior knowledge of the spatial configuration among the labelled organs to resolve spurious responses in the network outputs. Furthermore, we modified the architecture of the segmentation model to reduce its memory footprint as much as possible without drastically impacting the quality of the predictions. Lastly, we implemented a minimal inference script for which we optimized both execution time and required GPU memory.
♻ ☆ Read, Parse, Describe: Unified Document Parsing with Visual Element Description Generation NeurIPS 2026
Yuheng Chen, Yufan Chen, Zhuojun Cai, Ruiping Liu, Junwei Zheng, Jiale Wei, Di Wen, Weijia Fan, Kunyu Peng, Jiaming Zhang, Rainer Stiefelhagen
Document understanding systems are typically evaluated as separate components via layout analysis, optical text recognition, reading-order prediction, and image captioning. This separation leaves scientific figures, captions, and visual semantics weakly connected, and fails to measure whether a model can produce a single page-level parse that is both structurally faithful and semantically accessible. We introduce Read-Parse-Describe, a unified dataset for document parsing with visual element description generation. Given a page image, models are required to output an ordered list of layout elements containing bounding boxes, semantic labels, reading-order indices, recognized textual content, and alternative text descriptions for visual elements. The benchmark is built from 2,402 documents, comprising 18,163 continuous pages, 125,218 layout elements, and 11,338 visual instances. Besides, we further propose a plug-and-play space-awareness enhancer that reuses multi-layer visual features and injects grid-aware spatial guidance into vision-language tokens for layout-aware generation. Experiments across end-to-end VLMs and pipeline parsers show that existing systems still struggle to jointly recover reading order and visual semantics, while our enhanced model achieves strong joint performance. These results highlight the need to move beyond isolated document parsing modules toward unified models that jointly read, parse, and describe complex documents. The new dataset and code will be made publicly available at https://github.com/yuheng98chen/RPD-Unidoc
comment: Accepted by NeurIPS 2026
♻ ☆ Is the Geometry Doing the Work? An Operating-Point Audit of Hierarchy in Hyperbolic Vision-Language Models
Hyperbolic vision-language models are designed to encode abstraction geometrically: general concepts near the origin, specific ones farther out, and entailment cones representing directed order. We ask whether trained MERU, HyCoCLIP, and PHyCLIP models actually use these mechanisms. We audit seven released checkpoints and matched from-scratch interventions. All audited converged checkpoints remain near-Euclidean in the dimensionless radius $u=\sqrt{c}ρ$: the largest observed image-side value is $0.37$ -- well below $u\approx0.84$, where local metric distortion reaches $10\%$. Releasing the curvature floor changes curvature and norms but not this regime, with mixed, generally modest downstream shifts. Trained entailment cones are saturated or nearly saturated, so low violation rates can arise from trivially wide cones rather than learned order. Preregistered semantic traversal detects weak within-branch order but no operative full-hierarchy readout. Shuffle-controlled tests detect no pair-specific radial ordering in released checkpoints, and no positive result is consistent across all three matched ViT-B seeds. This non-detection is sensitivity-bounded: the diagnostic reaches $80\%$ power only at $\approx13.5$ pp of shuffle-controlled excess, while the largest released gap is $\approx5$ pp. We trace this to a low-curvature shortcut: lowering curvature widens entailment cones and suppresses violations without learning order. In the probed trajectories, gradient decomposition identifies entailment as the dominant curvature-lowering pressure during collapse. Yet curvature contracts even when entailment is removed, so the shortcut is not the sole cause. Under our diagnostics, the audited formulations do not demonstrate an operative radial or cone-based hierarchy. We distill the audit into a five-number geometry report for evaluating future hierarchy claims.
comment: 55 pages, 7 figures. Published in Transactions on Machine Learning Research (11/2026). Review: https://openreview.net/forum?id=Rj5ewBz62j
♻ ☆ The microscope is the mask: privileged views and labels from a cryo-ET forward model NeurIPS 2026
We explore the use of simulated data for training a model for protein annotation in crowded cryo-electron tomography volumes reconstructed from images collected at limited tilt angles and severely corrupted by the measurement operator. Firstly, we leverage the corruptions imposed by the forward model to generate domain-specific augmented paired views of the exact same scene for an invariance objective integrated into the LeJEPA self-supervised training framework. Secondly, we use additional information from the simulation pipeline such as the positions and identity of proteins in the simulated volumes to inform the architecture of the model and the loss function, so that semantic information is localised at protein positions in the resulting dense feature volume. The resulting model, CARNIVAL, is evaluated without finetuning on classification and detection tasks in real tomograms, using a benchmark dataset containing multiple protein types and two tomogram processing types. We show that CARNIVAL outperforms a state-of-the-art model trained using a contrastive objective on simulated data but without forward model-based paired views or privileged information.
comment: 17 pages, 7 figures. Paper accepted to NeurIPS 2026 workshop Sim2Science: ML with Imperfect Scientific Models
♻ ☆ TEMPO: Semantic-Action Decoupled RL Post-Training for Vision-Language-Action Models
Vision-language-action (VLA) models are commonly adapted to downstream manipulation tasks via supervised fine-tuning (SFT) or online reinforcement learning(RL) post-training. SFT is prone to distribution mismatch, and existing RL approaches typically optimize all trainable components within a single RL loop at a uniform update rate, although these components play distinct functional roles. We propose TEMPO, a semantic-action decoupled, two-timescale RL post-training framework for VLA models. TEMPO freezes the pretrained vision-language backbone and adapts two components with dedicated RL loops: the semantic projection layer, which produces the latent action conditioning the action expert, and the low-level action expert itself. The semantic projection layer is updated infrequently and the action expert frequently, which aims to limit latent-action drift while allowing the action expert to learn from online feedback. Our analyses show that, compared with equal update frequencies, less frequent semantic updates reduce latent-action drift and improve policy performance. Experiments on the CALVIN benchmark and real-world manipulation tasks demonstrate that TEMPO outperforms both pretrained state-of-the-art VLA models and the RL post-training baseline, while reaching and maintaining higher evaluation rewards on three real-world tasks.
♻ ☆ DPGS-SLAM: Diffusion-Prior and Provenance-Aware Monocular Gaussian Splatting SLAM for Dynamic Visual Sensing
This paper presents DPGS-SLAM (Diffusion-Prior and Provenance-Aware Monocular Gaussian Splatting SLAM) for dense mapping in environments containing dynamic objects. Objects in dynamic motion remain a significant challenge for dense visual SLAM since dynamic observations can contaminate camera tracking and scene reconstruction. Although 3DGS (3D Gaussian splatting) enables efficient dense mapping, existing 3DGS-SLAM systems may exhibit instability when dynamic objects violate their underlying static-scene assumption. Excluding dynamic observations can reduce their influence on tracking and mapping, but may leave occluded static backgrounds missing or weakly constrained. Our proposed DPGS-SLAM is a monocular RGB SLAM framework that supports provisional representations of insufficiently observed static backgrounds through asynchronous diffusion-based background completion and provenance-aware Gaussian mapping. Dynamic regions are identified through optical-flow-based motion discovery and temporally consistent mask propagation by SAM2. For selected keyframes, a fine-tuned diffusion inpainting model generates plausible background appearance in occluded regions, while metric depth predicted from completed images initializes provisional Gaussian primitives. Provenance labels distinguish generated primitives from observation-supported primitives, excluding generated content from camera-pose estimation and observed-map optimization. When previously occluded backgrounds become visible, eligible generated primitives are removed, and observation-supported primitives are independently initialized from real RGB observations and predicted depth. Experimental results demonstrated competitive camera tracking, with average ATE RMSE values of 1.2 cm on TUM RGB-D and 2.7 cm on BONN RGB-D datasets. On StaDy4D, DPGS-SLAM improved average PSNR by 4.60 dB over WildGS-SLAM using only monocular RGB input.
♻ ☆ AgroOmni: A Large-Scale Multi-view Agricultural Dataset for Cross-Scale Multimodal Reasoning
Jiarui Zhang, Junqi Hu, Zurong Mai, Yang Liu, Yuhang Chen, Shuohong Lou, Henglian Huang, Hong Cheng, Lingyuan Zhao, Jianxi Huang, Yutong Lu, Haohuan Fu, Juepeng Zheng
Modern agricultural data is sourced from diverse platforms and spans multiple spatial scales, ranging from ground-level close-up photography to Unmanned Aerial Vehicle (UAV) aerial observation and satellite remote sensing imagery. Accordingly, agricultural multimodal reasoning demands robust cross-scale spatial understanding. However, due to the lack of multi-view agricultural benchmark datasets, existing multimodal large language models (MLLMs) exhibit severe ground-level bias, which leads to scale confusion then semantic collapse in agricultural perception tasks, such as misinterpreting farmland imagery as walls or floors. To address this, we introduce AgroOmni, a large-scale multi-view training corpus with 268,119 Visual Question Answering pairs covering 56 specialized task categories across 14 task types, designed to capture diverse scales in modern precision agriculture.Built on this dataset, we propose AgroNVILA, which achieves a new state-of-the-art of 59.86% on the AgroMind benchmark (+12.57 over GPT-5.2), effectively mitigating the multi-view cross-scale gap for holistic agricultural understanding. Diagnostic evaluations on AgMMU further reveal an inherent heterogeneity between macro-priors and micro-diagnostics through constrained zero-shot performance. Meanwhile, even minimal fine-tuning leads to a dramatic performance gain of AgroNVILA on AgMMU, strongly demonstrating its generalization capability empowered by AgroOmni. Full training scripts are publicly available at https://github.com/Jrookie2005/AgroOmni, and the AgroOmni dataset is released at https://huggingface.co/datasets/AgroOmni/AgroOmni.
♻ ☆ VidEoMT: Your ViT is Secretly Also a Video Segmentation Model CVPR 2026
Narges Norouzi, Idil Esen Zulfikar, Niccolò Cavagnero, Tommie Kerssies, Bastian Leibe, Gijs Dubbelman, Daan de Geus
Existing online video segmentation models typically combine a per-frame segmenter with complex specialized tracking modules. While effective, these modules introduce significant architectural complexity and computational overhead. Recent studies suggest that plain Vision Transformer (ViT) encoders, when scaled with sufficient capacity and large-scale pre-training, can conduct accurate image segmentation without requiring specialized modules. Motivated by this observation, we propose the Video Encoder-only Mask Transformer (VidEoMT), a simple encoder-only video segmentation model that eliminates the need for dedicated tracking modules. To enable temporal modeling in an encoder-only ViT, VidEoMT introduces a lightweight query propagation mechanism that carries information across frames by reusing queries from the previous frame. To balance this with adaptability to new content, it employs a query fusion strategy that combines the propagated queries with a set of temporally-agnostic learned queries. As a result, VidEoMT attains the benefits of a tracker without added complexity, achieving competitive accuracy while being 5x-10x faster, running at up to 160 FPS with a ViT-L backbone. Code: https://github.com/tue-mps/videomt
comment: CVPR 2026. Code: https://github.com/tue-mps/videomt
♻ ☆ Open-Set Supervised 3D Anomaly Detection: An Industrial Dataset and a Generalisable Framework for Unknown Defects
Hanzhe Liang, Luocheng Zhang, Junyang Xia, HanLiang Zhou, Bingyang Guo, Yingxi Xie, Can Gao, Ruiyun Yu, Jinbao Wang, Pan Li
Although self-supervised 3D anomaly detection assumes that acquiring high-precision point clouds is computationally expensive, in real manufacturing scenarios it is often feasible to collect a limited number of anomalous samples. Therefore, we study open-set supervised 3D anomaly detection, where the model is trained with only normal samples and a small number of known anomalous samples, aiming to identify unknown anomalies at test time. We present Open-Industry, a high-quality industrial dataset containing 15 categories, each with five real anomaly types collected from production lines. We first adapt general open-set anomaly detection methods to accommodate 3D point cloud inputs better. Building upon this, we propose Open3D-AD, a point-cloud-oriented approach that leverages normal samples, simulated anomalies, and partially observed real anomalies to model the probability density distributions of normal and anomalous data. Then, we introduce a simple Correspondence Distributions Subsampling to reduce the overlap between normal and non-normal distributions, enabling stronger dual distributions modeling. Based on these contributions, we establish a comprehensive benchmark and evaluate the proposed method extensively on Open-Industry as well as established datasets including Real3D-AD and Anomaly-ShapeNet. Benchmark results and ablation studies demonstrate the effectiveness of Open3D-AD and further reveal the potential of open-set supervised 3D anomaly detection.
comment: The submitter did not have the rights to agree to the license at the time of submission.
♻ ☆ ENIGMA-360: An Ego-Exo Dataset for Human Behavior Understanding in Industrial Scenarios
Francesco Ragusa, Rosario Leonardi, Michele Mazzamuto, Daniele Di Mauro, Camillo Quattrocchi, Alessandro Passanisi, Irene D'Ambra, Antonino Furnari, Giovanni Maria Farinella
Understanding human behavior from complementary egocentric (ego) and exocentric (exo) points of view enables the development of systems that can support workers in industrial environments and enhance their safety. However, progress in this area is hindered by the lack of datasets capturing both views in realistic industrial scenarios. To address this gap, we propose ENIGMA-360, a new ego-exo dataset acquired in a real industrial scenario. The dataset is composed of 180 egocentric and 180 exocentric procedural videos temporally synchronized offering complementary information of the same scene. The 360 videos have been labeled with temporal and spatial annotations, enabling the study of different aspects of human behavior in industrial domain. We provide baseline experiments for 4 foundational tasks for human behavior understanding: 1) Temporal Action Segmentation, 2) Keystep Recognition, 3) Egocentric Human-Object Interaction Detection, and 4) Video-Grounded Cross-View VQA, showing the limits of state-of-the-art approaches on this challenging scenario. These results highlight the need for new models capable of robust ego-exo understanding in real-world environments. We publicly release the dataset and its annotations at https://fpv-iplab.github.io/ENIGMA-360/.
♻ ☆ A Marine Debris Detection Framework for Ocean Robots via Self-Attention Enhancement and Feature Interaction Optimization
Marine debris detection for ocean robot is crucial for ecological protection, yet performance is often degraded by low-quality images with blur, complex backgrounds, and small targets. To address these challenges, we propose YOLO-MD, an enhanced YOLO-based detection framework. A Dual-Branch Convolutional Enhanced Self-Attention (DB-CASA) module is designed to strengthen spatial-channel interactions, improving feature representation in degraded images. Additionally, a lightweight shift-based operation is introduced to enhance fine-grained feature extraction for objects of varying scales while maintaining parameter efficiency. We further propose SFG-Loss to mitigate class imbalance and optimization instability via dynamic sample reweighting. Experiments on the UODM dataset demonstrate that YOLO-MD achieves 0.875 precision, 0.822 F1-score, and 0.849 mAP50, outperforming the latest state-of-the-art methods. The effectiveness of this method has also been verified through real-world robotic edge deployment experiments.
comment: Language polishing only; no substantive changes to content, data, results, or conclusions
♻ ☆ Learning to Live: A Step Toward Lifelong Embodied Intelligence
Lirong Che, Peidong Wang, Yimeng Su, Jiong Wang, Yang Chen, Wenbo Fu, Zhenyu Huang, Tiancheng Xu, Xueqian Wang, Yuzhe Yang, Dongli Xie, Xu Cao, Chuang Wang, Daqi Liu, Jian Su
Embodied agents are usually organized around individual tasks, yet the physical world keeps changing between tasks and during execution. We introduce LIFE (Lifelong Intelligence from Experience), a framework in which a robot observes continuously, remembers selectively, and turns experience into new capabilities. RoboBeats, a family of lightweight streaming gates, tracks conditions set by the agent and discovers new events, deciding when an observation warrants a response or a memory write. Selected observations enter Plastic Memory, a flat store of images and structured records that keeps the source evidence, so new fields and indexes can be added when later questions need them. When existing tools fall short, Explore-Distill-Mount (EDM) turns solutions found during problem solving into validated, reusable tools and skills. Under matched protocols, RoboBeats substantially improves condition tracking over its base models and, at the same writing budget, retains more useful evidence for memory question answering than uniform sampling. A case study of limited scale, replaying 24 hours of continuous video recorded by a real robot, illustrates how the integrated system builds memory from a day of experience and uses it to answer everyday questions. LIFE takes a step toward robots that become more familiar with their surroundings and more capable through experience: learning not only to complete tasks, but to live in the world.
♻ ☆ SimpleSearch-VL: A Simple Recipe for Multimodal Agentic Deep Search
Ming Dai, Zhihong Lu, Jinjie Gu, Jiedong Zhuang, Jing Hao, Lijun Wu, Yefeng Liu, Wankou Yang, Jian Wang, Chunhua Shen
We present SimpleSearch-VL, an efficient, reliable, and practical framework for multimodal agentic search. Its core idea is to improve the agent's own search-and-verification process rather than scaling data, tools, or auxiliary model components. For efficiency, Factorized Adaptive Rollout (FAR) improves sampling efficiency by forming more informative training groups while using redundant samples to mitigate long-tail latency and expose hard samples. For reliability, SimpleSearch-VL performs evidence-verified reasoning, explicitly using chain-of-thought verification to assess the relevance of retrieved visual and textual cues to the original context. For practicality, SimpleSearch-VL keeps a lightweight tool interface and performs webpage self-summary within the agent, requiring no additional external dependencies. With only 5K supervised tool-interleaved trajectories and 2K RL data, SimpleSearch-VL improves Qwen3-VL agentic baselines by 15.8 and 16.0 average points for the 8B and 30B-A3B variants, respectively. The SimpleSearch-VL-30B-A3B model further achieves performance competitive with agentic Gemini-3-Pro.
comment: Technical Report
♻ ☆ HAVANA: A Hierarchical Video-Action Model for Vision-Language Navigation
Vision-Language Navigation (VLN) requires long-horizon instruction-consistent planning and short-horizon path planning grounded in the currently observed geometry. Anticipating how future observations evolve along the instructed route provides predictive guidance beyond the current field of view, motivating video-action models (VAMs) that learn representations of future visual states. However, many existing VAMs couple future visual prediction with action generation during inference, despite their distinct temporal requirements: instruction-conditioned route guidance can be reused across multiple control steps, whereas local trajectories must continuously adapt to fresh observations. To bridge this gap, we introduce HAVANA, a dual-system VAM that decouples predictive reasoning from reactive local planning. A slow system produces a compact language-conditioned latent plan from observation history, while a fast system combines the latest plan with the current observation to generate local trajectories. Their asynchronous execution enables rapid adaptation without repeated future prediction, and no future video is generated during inference. Experiments on OmniNavBench demonstrate state-of-the-art navigation performance over prior baselines, while zero-shot real-world deployments further show robust generalization to unseen environments.
comment: 10 pages,6 figures
♻ ☆ InstantHMR : Real-Time Human Mesh Recovery on Your Phone
Real-time 3D human mesh recovery from a single image on mobile devices is a challenging problem with many applications, including robotics, biomechanics, AR/VR, and exergames. Current state-of-the-art models are either too slow to run efficiently on mobile devices or sacrifice accuracy for speed. We present InstantHMR, a family of open-source single-image regressors suited for mobile inference, based on the Momentum Human Rig (MHR) parametric model and trained exclusively on publicly available annotations from the SAM 3D Body dataset. Our approach achieves remarkable performance on a Samsung Galaxy S23 NPU, recovering a 6,890-vertex body mesh in 5.5 ms per person, and 0.98 ms on an NVIDIA RTX 4070 GPU. This speed surpasses all existing methods we evaluated. On mobile, it is 12x faster than the single mobile-compatible method that slightly outperforms ours in accuracy, while achieving competitive accuracy with 40.0 mm PA-MPJPE on 3DPW, 48.8 mm on EMDB and 41.7 mm on RICH. To ensure fair comparisons, we re-evaluate 11 state-of-the-art methods under a unified protocol aligned with the closest MHR-based competitors, including Meta's official MHR-to-SMPL conversion, and report their reproduction results. Additionally, we measure latency across eight different targets, ranging from desktop TensorRT to two phone NPUs. Code and models are available at https://github.com/Naturalpad-studio/InstantHMR.
♻ ☆ Medical Image Segmentation based on Deep Active Contour and Mean Curvature Loss Function
Pixel-level supervision in medical image segmentation ignores the geometric priors of the target region. Chan-Vese losses restore region-level priors and boundary regularity but do not characterise local boundary geometry. We propose the Deep Active Contour and Mean Curvature (DACMC) loss, which embeds the mean curvature of the predicted boundary in the Chan-Vese region functional; a fixed fourth-order-accurate convolution kernel evaluates curvature without explicit differentiation. The loss adds no trainable parameters and no inference cost, increases training time by 4.8%, and uses one hyper-parameter lambda = 10^-3 shared by all settings. We evaluate DACMC on liver CT, spleen MRI and ISIC dermoscopy images with U-Net and CE-Net backbones, four metrics (DSC, HD95, JS, ASD) and four competing losses under an identical protocol, together with a five-point lambda sweep, a wall-clock cost measurement and a five-seed replication with paired Wilcoxon signed-rank tests. DACMC ranks first or second in Dice in five of the six dataset-backbone settings, most consistently in distance metrics. On spleen MRI it reduces HD95 and ASD to 16.28 and 1.90 mm, and on ISIC dermoscopy it reduces HD95 to 7.08 mm. Dice varies by 0.0023 across five lambda values spanning three orders of magnitude; degradation appears only when curvature dominates. Five-seed replication shows that curvature stabilises training and removes occasional severe failures, although mean differences from the closest competing loss do not reach statistical significance at five seeds. DACMC imposes second-order geometric regularity on segmentation boundaries at low overhead, with the clearest benefit for small or scale-varying targets. Limitations include dataset size and partitioning, classical backbones and losses, and restriction to 2-D single-organ images.
comment: Revised version: restructured the abstract, added a five-seed replication with paired statistical tests and a discussion of limitations, unified the cross-entropy notation, and updated the reference list and author declarations. 39 pages, 7 figures, 7 tables
♻ ☆ FACE: Few-shot Adapter with Cross-view Fusion for Cross-subject EEG Emotion Recognition
Cross-subject EEG emotion recognition is challenged by significant inter-subject variability and intricately entangled intra-subject variability. Existing works have primarily addressed these challenges through domain adaptation or generalization strategies. However, they typically require extensive target subject data or demonstrate limited generalization performance to unseen subjects. Recent few-shot learning paradigms attempt to address these limitations but often encounter catastrophic overfitting during subject-specific adaptation with limited samples. This article introduces the few-shot adapter with a cross-view fusion method called FACE for cross-subject EEG emotion recognition, which leverages dynamic cross-view fusion and effective subject-specific adaptation. Specifically, FACE incorporates a cross-view fusion module that dynamically integrates global brain connectivity with localized patterns via subject-specific fusion weights to provide complementary emotional information. Moreover, the few-shot adapter module is proposed to enable rapid adaptation for unseen subjects while reducing overfitting by enhancing adapter structures with meta-learning. Experimental results on three public EEG emotion recognition benchmarks demonstrate the effectiveness of FACE under few-shot cross-subject evaluation.
comment: Early Access Version
♻ ☆ Geometry Matters in Packed 3D Attention NeurIPS 2026
Packing a 3D field into a token sequence gives every token two indices: a coordinate in the sample and a position in memory. Raster, Morton, and block-contiguous layouts change the position and leave the coordinate alone. A positional mechanism indexed by sequence position can therefore learn the memory layout rather than the object, and accuracy on the training layout cannot reveal that it has. We introduce PACK3D, a local-global attention operator that separates memory layout from attention geometry. It attends exactly within a coordinate-defined neighborhood, sees the whole field through pooled summaries, and rotates queries and keys with axial RoPE at token-grid coordinates. Because keys are rotated before pooling, each query-to-summary logit is a weighted sum of query-key terms evaluated at the summarized tokens' displacements from the query. The operator is equivariant to any repacking that carries coordinates, neighborhoods, summary memberships, and within-block slots along with the tokens. When evaluated on Allen WTC-11 microscopy at 262k tokens, PACK3D and a matched packed-sequence RoPE control reach comparable Dice on the training layout. Repacking the same validation volumes then costs the packed control 0.048 macro Dice on average, with a loss in all six seeds, while PACK3D changes by 0.0001. On MSD Heart and Spleen, sequence indexing again changes more predictions under repacking, at little cost in Dice. PACK3D also exceeds a parameter-matched 3D U-Net by 0.009 Dice over six seeds, and a full PACK3D layer runs within 7-31% of window-attention latency while adding a global pathway. Geometry, not storage order, should index attention over packed 3D fields. Code is available at https://github.com/amilworks/Pack3D.
comment: Accepted to NeurIPS 2026 (Poster)
♻ ☆ PhysPatch: A Physically Realizable and Transferable Adversarial Patch Attack for Multimodal Large Language Models-based Autonomous Driving Systems
Multimodal Large Language Models (MLLMs) are becoming integral to autonomous driving (AD) systems due to their strong vision-language reasoning capabilities. However, MLLMs are vulnerable to adversarial attacks, particularly adversarial patch attacks, which can pose serious threats in real-world scenarios. Existing patch-based attack methods are primarily designed for object detection models and perform poorly when transferred to MLLM-based systems due to the latter's complex architectures and reasoning abilities. To address these limitations, we propose PhysPatch, a physically realizable and transferable adversarial patch framework tailored for MLLM-based AD systems. PhysPatch jointly optimizes patch location, shape, and content to enhance attack effectiveness and real-world applicability. It introduces a semantic-based mask initialization strategy for realistic placement, an SVD-based local alignment loss with patch-guided crop-resize to improve transferability, and a potential field-based mask refinement method. Extensive experiments across open-source, commercial, and reasoning-capable MLLMs demonstrate that PhysPatch significantly outperforms prior methods in steering MLLM-based AD systems toward target-aligned perception and planning outputs. Moreover, PhysPatch consistently places adversarial patches in physically feasible regions of AD scenes, ensuring strong real-world applicability and deployability.
♻ ☆ CamChoreo: Benchmarking Temporally Grounded Compositional Camera Motion Understanding for Vision Language Models
Dazhao Du, Shiyan Du, Jian Liu, Yongjian Yu, Bohai Gu, Tao Han, Jie Zhang, Hualuo Liu, Eric Liu, Yujia Zhang, Xi Chen, Song Guo
Understanding camera motion is fundamental to video perception, with applications in spatial intelligence and controllable video generation. Existing camera motion evaluations typically assign labels to entire clips, overlooking within-shot changes and simultaneous movements. We therefore formulate camera motion understanding as temporally grounded, compositional recognition, requiring models to localize motion-consistent intervals and identify all active movements. We introduce CamChoreo, the first benchmark for this task, comprising 4,229 diverse single-shot clips with expert temporal annotations using 20 direction-aware labels. Nearly half of its segments contain compound camera motion. We also construct a separate training corpus of approximately 43K clips with the same annotation format. On CamChoreo, supervised fine-tuning (SFT) on this corpus achieves 62.1/64.1 micro F1 with 4B/8B backbones, outperforming advanced closed-source vision language models (VLMs). To complement the primarily semantic representations of VLM visual encoders with cross-frame geometry, we introduce CamInject, which injects features from a frozen 3D foundation model into the VLM. CamInject improves over SFT but requires running the costly 3D model at inference. We instead propose CamDistill, which distills the same geometric knowledge into lightweight camera tokens during training and removes the 3D model at inference. CamDistill improves micro F1 over same-data SFT by 5.4/3.7 points on 4B/8B backbones, closely approaching CamInject. CamDistill-8B also demonstrates strong zero-shot transfer on the external camera motion understanding benchmarks CameraBench and CameraMotionVQA. Together, these resources and CamDistill enable benchmarking and improving temporally grounded, compositional camera motion understanding.
♻ ☆ UniFlow-RS: Unified Generative Prediction for Heterogeneous Remote-Sensing Tasks
Remote-sensing foundation models increasingly unify representations across sensing modalities and downstream tasks, yet their prediction mechanisms remain largely task-specific. Separate decoders, prediction heads, and objectives leave prediction fragmented across tasks. This naturally raises a question: can heterogeneous remote-sensing tasks share the prediction process? UniFlow-RS answers this question with a single latent flow-matching predictor shared across tasks. This requires addressing two key challenges: unifying heterogeneous prediction spaces and harmonizing joint optimization across tasks. First, we cast heterogeneous task outputs into structured visual targets in a common latent space, allowing them to be modeled by the same pretrained flow predictor. Task-dependent conditioning further accommodates differences in input configurations. Second, we optimize all tasks with a common Flow Matching objective, eliminating task-specific prediction losses. To mitigate optimization mismatch across tasks and modalities, we introduce Task-Modality Adaptive Timestep Sampling (TMAS), which dynamically allocates training timesteps according to the online learning state of each task-modality group. Using a single shared model, UniFlow-RS handles six tasks and thirteen datasets, surpassing recent task-specific state-of-the-art methods on several benchmarks while remaining competitive on the others. Beyond these benchmarks, the same checkpoint demonstrates strong zero-shot transfer across unseen datasets, task settings, occlusion conditions, and near-infrared false-color imagery, without target-specific adaptation. Codes are available at https://github.com/MindVision-AI/UniFlow-RS.
♻ ☆ Interactive Visual Gym: Evaluating Visual Interaction Models with Simulated Users in Interactive Worlds
Chenhao Zheng, Zixian Ma, Kai Li, Kanchana Ranasinghe, Jieyu Zhang, Artemis Panagopoulou, Weikai Huang, Jianxu Shangguan, Yuanyuan Tan, Silvio Savarese, Juan Carlos Niebles, Ranjay Krishna, Ran Xu, Honglu Zhou
Visual interaction models that support users during ongoing physical tasks require evaluation in closed loop, where assistant responses can change subsequent user behavior and task outcomes. Existing benchmarks largely rely on offline replay and therefore cannot capture these effects. We introduce Interactive Visual Gym, a framework for closed-loop evaluation and training of visual interaction models through joint simulation of executable task environments and human users. We contribute three simulators that recreate everyday tasks and human-aligned action spaces across cooking, household manipulation, and mobile-interface domains, together with a persona-conditioned user simulation harness that collaborates with the assistant while incorporating controlled execution errors and other out-of-plan behaviors. We instantiate the framework as IVG-Bench, a benchmark with 315 interaction scenarios. Evaluating state-of-the-art VLMs as visual interaction models reveals substantial limitations: the best-performing model, Claude Opus 5.5, completes only 53.5% of tasks within the time limit, exhibits significant weaknesses in factual grounding and actionable guidance, and is often too slow to prevent errors when response latency is considered. We verify our evaluation protocol with a human study which finds 78.5% agreement between our interaction-quality metric and participant preferences. We further show that the environments provide useful learning signals: in a proof-of-concept GRPO experiment, closed-loop training improves both task success and interaction quality for Qwen3.5-4B model in our cooking environment.
comment: 39 pages, including references and appendices
♻ ☆ AVC-Reward: Event-Centric Audio-Video Consistency Evaluation
As joint audio-video generation models become more capable, they can generate audio-visual scenes exhibiting three forms of complexity: multi-event composition, multi-layer audio mixture, and asymmetric cross-modal relations. However, current evaluations largely reuse metrics from earlier simple and unidirectional tasks. They typically evaluate an audio-video clip as a whole and rely on simple predefined relations, making them unsuitable for these increasingly complex scenes. We propose an event-centric evaluation framework that disentangles audio and visual events at the event level, analyzes each event independently, cross-validates relations across modalities, and aggregates the judgments into a coarse-to-fine, three-level consistency metric. The metric assesses global narrative harmony, event-level temporal synchronization, and fine-grained attribute alignment, providing a general formulation for audio-video consistency evaluation across diverse generation settings and cross-modal relations. We instantiate the metric as AVC-Reward, a trainable audio-video reward model. To evaluate the agreement between automatic judgments and human perception across generation paradigms and relation complexities, we construct AVC-Benchmark, which is primarily built from diverse positive and negative samples generated by various types of audio-video generation models. AVC-Reward achieves state-of-the-art alignment with human judgment among automatic methods, provides interpretable rationales that localize failures, and improves the audio-video consistency of a joint generation model when used as an online reinforcement-learning signal. Project page: https://tsuki0512.github.io/AVC-Reward.github.io/
♻ ☆ CAT-Free: Calibration-, Annotation-, and Target-Training-Free Multi-View BEV Pedestrian Detection and Tracking
Multi-camera pedestrian detection and tracking often require camera calibration, position annotations, and target-scene training. CAT-Free removes all three requirements and uses only synchronized RGB video. From this video, CAT-Free estimates camera poses and the ground plane. With the estimated poses and ground plane, CAT-Free triangulates detections across cameras and links pedestrian positions into tracks. However, camera errors and mismatched detections produce false positions. Two geometric filters remove these false positions: one rejects unstable estimates, and the other rejects positions off the ground. Both filter thresholds are estimated from the unlabeled video. Without calibration, annotations, or target-scene training, CAT-Free achieves 89.2, 84.1, and 76.1 MODA on WildTrack, MultiviewX, and GMVD. CAT-Free is also less affected by camera errors than methods that assume known camera parameters, and it outperforms a domain adaptation method without any adaptation.
♻ ☆ Anatomy-Decomposed Chest Computed Tomography (CT) Projections as Scalable Supervision for Bone Suppression in Chest Radiographs
Mrunmay Angaitkar, Piyush Kumar, Aarjav Satia, Pranav Rao, Ashish Mittal, Manoj Tadepalli, Preetham Putha
Bone overlap can obscure abnormalities in chest radiographs, while scarce paired training data limit supervised bone suppression. We address this challenge with a digitally reconstructed radiograph (DRR) framework that converts chest computed tomography (CT) into paired supervision for component suppression. A novel bone segmentation algorithm enables CT decomposition into bone, non-lung soft-tissue, and lung components, which are projected separately. Their weighted combination yields synthetic radiographs with pixel-registered component images that sum exactly to the full DRR. Models trained on these data suppress bone or lung components by predicting the target component and recovering the remainder by subtraction, transferring to real radiographs without real paired training data. As an extension, their outputs on real radiographs provide target domains for unpaired, component-wise DRR translation, reducing the appearance gap while retaining anatomical details. Across multiple public datasets, downstream detection experiments demonstrate the utility of bone suppression, with gains concentrated on abnormalities with substantial bone overlap. Compared with open-source DRR engines applied to the same CTs, our unmodified DRRs achieve comparable realism and preservation of label-relevant anatomy, while translated DRRs achieve the best Fréchet inception distance (FID), lung-field sharpness, and agreement with source-CT anatomy among the evaluated methods. Models and inference code: https://huggingface.co/qureaiorg/bone-suppression ; Translated projections: https://huggingface.co/datasets/qureaiorg/ct2xr-projections.
♻ ☆ RAIN: Region-Aware Inversion Network for Semantic Watermark Extraction BMVC 2026
Semantic watermarks for diffusion models embed ownership information into the generative process while preserving perceptual quality, but Gaussian-Shading extraction conventionally requires multi-step diffusion inversion to recover the initial noise. Recent one-step methods show that this cost can be reduced substantially. We study this problem through extended flow matching and conditional regression. The key observation is that, near the high-SNR image endpoint, recovering a useful noise statistic given by the first-step output of the extended flow matching in the high-SNR regime is much simpler than reconstructing the full inverse trajectory, and Gaussian Shading only requires the recovered latent to remain in the correct watermark decision region. Based on this observation, we propose a lightweight, prompt-free extractor that decomposes endpoint recovery into an image-like anchor and a noise-oriented residual, which increases the capability of the model to utilize GPU parallel computation. The resulting method avoids iterative inversion and repeated evaluation of a diffusion-scale U-Net, providing an efficient one-step extraction pipeline with a concise theoretical interpretation. The computational cost of extracting noise is lower than that of both OSI and FARI. The github repo is there: https://github.com/TheLovesOfLadyPurple/RAIN-lightweight-NN-for-one-step-semantic-watermark-extraction
comment: Accepted to the BMVC 2026 Workshop (TrustVis) as an Oral Presentation
♻ ☆ CORAL: A Benchmark for Structure-aware and Brain-wide Neuron Reconstruction in Light Microscopy
Automatic neuron reconstruction from light microscopy images is a central problem in computational neuroanatomy. While recent methods have achieved encouraging results on local image blocks, it remains unclear whether such progress translates to reconstruction that is both structurally accurate and scalable to the whole-brain scale. We present CORAL, the first benchmark for structure-aware evaluation of automatic neuron reconstruction from light microscopy images at both local and whole-brain scales. Built on a high-quality whole-brain fMOST dataset with carefully curated annotations, CORAL establishes two progressive tasks: block-level reconstruction, which evaluates reconstruction methods under limited spatial context, and brain-wide reconstruction, which assesses complete neuron reconstruction at the whole-brain scale. To account for topological correctness beyond geometric distance similarity, we introduce a structure-aware metric based on fiber prediction. To further achieve complete neuron reconstruction across the entire brain, we develop a brain-wide neuron tracing framework that extends arbitrary local reconstruction methods to the whole-brain scale through an iterative local-to-global process. Using this benchmark, we provide the first structure-aware comparison of mainstream methods for local neuron reconstruction and further evaluate their performance in brain-wide reconstruction. Our results underscore the importance of structure-aware evaluation and the need for more robust methods for complete neuron reconstruction.
♻ ☆ D2D: Detector-to-Differentiable Critic for Improved Numeracy in Text-to-Image Generation NeurIPS 2026
Text-to-image (T2I) diffusion models have achieved strong performance in semantic alignment, yet they still struggle with generating the correct number of objects specified in prompts. Existing approaches typically incorporate auxiliary counting networks as external critics to enhance numeracy. However, since these critics must provide gradient guidance during generation, they are restricted to regression-based models that are inherently differentiable, thus excluding detector-based models with superior counting ability, whose count-via-enumeration nature is non-differentiable. To overcome this, we propose Detector-to-Differentiable (D2D), a novel framework that transforms non-differentiable detection models into differentiable critics, leveraging their superior counting ability to guide numeracy generation. Specifically, we design custom activation functions to convert detector logits into soft binary indicators, which are then used to optimize the noise prior at inference time with pre-trained T2I models. Our extensive experiments on SDXL-Turbo, SD-Turbo, and Pixart-DMD across four benchmarks of varying complexity (low-density, high-density, and multi-object scenarios) demonstrate consistent and substantial improvements in object counting accuracy (e.g., boosting up to 13.7% on D2D-Small, a 400-prompt, low-density benchmark), with minimal degradation in overall image quality and computational overhead. Project page: https://nobliney.github.io/d2d/.
comment: Accepted to NeurIPS 2026. Camera-ready version
♻ ☆ Rethinking Floorplan Generation with Space Syntax: From Design Theory to Model Improvement
Architectural design increasingly adopts generative AI (GenAI) for floorplan generation, an early-stage design task exploring how spaces should be organized for use. Yet fidelity objectives inherited from native visual and geometric generators do not automatically capture this design purpose. This organization remains largely implicit in generated representations, limiting direct evaluation and feedback. Drawing on space syntax, an established architectural design theory, the Space Syntax Integration Operator (SSIO) recovers convex-space access graphs across heterogeneous formats. This theory-grounded shared intermediate representation, which we call a design container, supports evaluation and model understanding of spatial design. Applying SSIO to four residential corpora spanning 15 countries reveals public spaces' integration levels are highest, with corpus-dependent distributions. These patterns underpin the RPLAN-SSI benchmark, an empirical design reference system assessing population-level functional-profile alignment (IPD) and distributional alignment of within-plan public-space advantage (PIAD). Comparison with this residential reference reveals a design gap: public spaces in generated floorplans have a smaller integration advantage over other spaces. To address this gap, Space Syntax-Guided Post-Training (SSPT) provides a plug-and-play design-learning interface: SSIO-derived RL feedback reduces IPD and PIAD across five heterogeneous diffusion backbones while preserving native mechanisms. We further discuss how this approach may inform GenAI applications in other design domains.
comment: Revised and expanded version;
♻ ☆ NVIDIA OmniDreams: Real-Time Generative World Model for Closed-Loop Autonomous Vehicle Simulation
Aarti Basant, Amlan Kar, Despoina Paschalidou, Fangyin Wei, Francesco Ferroni, Guillermo Garcia Cobo, Haithem Turki, Huan Ling, Jaewoo Seo, James Lucas, Jay Zhangjie Wu, Jialiang Wang, Jonathan Lorraine, Jun Gao, Kai He, Katarina Tothova, Kevin Xie, Michal Tyszkiewicz, Qi Wu, Riccardo de Lutio, Ruilong Li, Sanja Fidler, Seung Wook Kim, Tianchang Shen, Tianshi Cao, Tobias Pfaff, William Lew, Xindi Wu, Xuanchi Ren, Yifan Lu, Yuxuan Zhang, Zan Gojcic, Zian Wang
As autonomous vehicle capabilities advance, the safe evaluation of driving policies in long-tail scenarios remains a critical bottleneck. In closed-loop simulation, the driving policy model actively interacts with the environment, where its actions dynamically update the simulator state and directly influence the next set of generated sensor observations. While recent reconstruction-based neural simulators offer photorealism, they are fundamentally constrained by their initial captured data and struggle to generalize to highly dynamic or novel scenes. To overcome these limitations, we introduce OmniDreams, a foundation generative world model mid- and post-trained from the Cosmos diffusion model to autoregressively generate action-conditioned videos in real time. By leveraging the rich visual priors of Cosmos and mid- and post-training on 21k hours of driving scenarios, OmniDreams synthesizes complex, unobserved phenomena that are hard for traditional simulators to capture, such as extreme weather and unpredictable dynamic agent behaviors. Crucially, it autoregressively conditions its photorealistic sensor generation on past frames, the current simulator state, and immediate driving actions. Deployed in a closed-loop system with the Alpamayo 1 policy model and AlpaSim orchestrator, OmniDreams acts as a highly responsive, reactive environment, providing a scalable and comprehensive solution for training and evaluating next-generation autonomous driving policies. We additionally show preliminary results indicating that a world-action model (WAM) post-trained from OmniDreams achieves strong performance on the Physical AI Autonomous Vehicles NuRec dataset, surpassing the VLA-based Alpamayo 1.5 research policy model while using only 1/5 the total parameters. These results highlight the potential for a real-time world model like OmniDreams to also serve as a backbone for policy architectures.
comment: Research blog: https://research.nvidia.com/labs/sil/projects/omnidreams-blog/, GitHub: https://github.com/nv-tlabs/omni-dreams, Model weights: https://huggingface.co/nvidia/omni-dreams-models
♻ ☆ When can multi-fragment 2D/3D registration be trusted? Calibrated certification and uncertainty-driven view planning for pelvic fracture fluoroscopy
Registering a segmented CT to fluoroscopy can locate every fracture fragment but can guide surgery only when trusted. After each shot, our closed loop registers all fragments, bounds their errors with a calibrated sandwich covariance, flags implausible or ambiguous ones, and stops or plans the next view. On simulated fluoroscopy of 60 held-out fracture CTs with realistic image uncertainty, 95% bounds covered 83% of unflagged fragments and 91% of those within 5mm, but from the second view on 14% failed by over 5mm, a third with bounds of at most 2mm. A fragment's shift when a view was added separated tightly bounded failures best (exploratory; AUC 0.94, bound 0.81). Requiring that the newest view moved no fragment beyond its bound, preferred post hoc, lowered false stops at 2mm from 23% to 6% overall, with 345 instead of 362 correct ones but more trajectories uncertified; delaying the stop did not, and other signals at equal acceptance left 10-15% false. With a segmentation error shared by all views (median Dice 0.89), 21% of this rule's stops were false against 10% without it (15 cases; exploratory). Minimax planning certified sooner than the clinical protocol without the check (4.20 versus 4.72 shots, Holm p <0.001), but no criterion reached the pre-specified endpoint, a true 2mm error, significantly sooner, even in the true basin. On real fluoroscopy of five intact pelves no 2mm certificate was false. Trust takes evidence across views, not only a calibrated local bound.
♻ ☆ SurgStreamer: proactive, request-driven monitoring of surgical video streams with structured state and a reusable reasoning layer
Surgical teams could benefit from assistants that watch the video for requested events, such as "tell me when the clip applier appears during Calot triangle dissection", and answer questions about what has already happened. We formalise this task as request-driven event detection with exact semantics and causal alert decisions, and present SurgStreamer. It separates procedure-specific perception, which maps the 1-FPS stream to a structured state over tools, actions and phases, from procedure-agnostic reasoning, which evaluates a formal monitoring condition over this state and an event memory, either with exact rules or with a small controller that addresses entities through pointers rather than names. We evaluate this monitoring core with gold conditions. On a pre-registered, sealed test set of 12 laparoscopic cholecystectomies, SurgStreamer reached a macro event-F1 of 0.598 over full procedures. On identical 75-s windows with only the window available to every method, it scored 0.677. The tested configurations of the streaming vision-language model OneStreamer-4B, including ones fine-tuned on SurgStreamer's state or on video with an adapted vision encoder, scored at most 0.540 (Holm-adjusted p <= 0.012). At three external centres, re-training only perception raised performance on held-out procedures from 0.239 to 0.470. On a new procedure, pituitary surgery, controller weights trained on cholecystectomy matched an in-domain controller on oracle state (0.947 vs 0.938). On predicted state they missed the pre-registered non-inferiority margin (-0.024 [-0.040, -0.008]) unless the alert thresholds were re-calibrated. The results support separating perception from reasoning and point to perception under distribution shift as the main limitation.
♻ ☆ Expanding the Role of Diffusion Models for Robust Classifier Training
Incorporating diffusion-generated synthetic data into adversarial training (AT) has been shown to substantially improve the training of robust image classifiers. In this work, we extend the role of diffusion models beyond merely generating synthetic data, examining whether their internal representations, which encode meaningful features of the data, can provide additional benefits for robust classifier training. Through systematic experiments, we show that diffusion models offer representations that are both diverse and partially robust, and that explicitly incorporating diffusion representations as an auxiliary learning signal during AT consistently improves robustness across settings. Furthermore, our representation analysis indicates that incorporating diffusion models into AT encourages more disentangled features, while diffusion representations and diffusion-generated synthetic data play complementary roles in shaping representations. Experiments on CIFAR-10, CIFAR-100, and ImageNet validate these findings, demonstrating the effectiveness of jointly leveraging diffusion representations and synthetic data within AT.
♻ ☆ Appear2Meaning: A Cross-Cultural Benchmark for Structured Cultural Metadata Inference from Images
Yuechen Jiang, Enze Zhang, Md Mohsinul Kabir, Chenying Lin, Ying Peng, Fengyuan Liu, Qianqian Xie, Stavroula Golfomitsou, Konstantinos Arvanitis, Sophia Ananiadou
Museum records express how people understand heritage objects through cultural and historical relations. Recovering these human attributions from images is a technical challenge for vision-language models (VLMs). We introduce Appear2Meaning, a cross-cultural benchmark of 750 museum objects for image-only recovery of title, culture, period, origin, and creator. Its evaluation framework distinguishes field-level from complete-record compatibility and uses human validation for semantic judgement. Across nine VLMs, mean field accuracy is 36.16--49.82%; 7.20--17.02% of records have all four contextual fields compatible, and 2.13--7.87% have all five, including title. For the two tested models, image-only input yields 12.6--32.2 percentage points higher reference compatibility for culture, period, and origin than a museum-and-title-only control.