Roxanne De Maeyer, Ine Dirks, Yannick Brackenier, Simon DeKeyser, Jef Vandemeulebroucke, Vicky De Ridder
The HECKTOR 2026 challenge benchmarks machine learning for head and neck cancer, where teams delineate primary and nodal tumour volumes, perform TN staging, and predict recurrence-free survival (RFS). This report describes the submissions for the HECKTOR challenge of team nucli. As the three tasks span both dense and global information from multicentric PET/CT scans, we utilise self-supervised pre-trained encoders, training a 2D DINOv2 encoder (PET-DINOv2-MIPs) on HECKTOR 2026's PET scans. Our final test submission ensembles a DynUNet trained from scratch with a SwinUNETR whose Swin trunk was initialised from a SimMIM-pre-trained checkpoint for segmentation, reaching a mean Dice of 0.5409. For TN staging, GTVp/GTVn geometric features from the segmentation masks are fit to stage probabilities by logistic regression (LR) and fused with PET-DINOv2-MIPs deep features (PCA + LR) via a second-level LR, reaching balanced accuracies of 0.3962 (T-stage) and 0.6432 (N-stage). For the RFS prognosis task, we fine-tune the last two blocks of the PET-DINOv2-MIPs ViT-S encoder jointly with an MLP head under a CoxPH loss, concatenating clinical covariates with the imaging features, reaching a test Harrell's C-index of 0.6577 and a weighted challenge score of 0.5802. The prognosis result is comparable to the best entry of the previous edition and consistent with our internal cross-validation, while our segmentation result is well below it and limits the staging results that depend on it.
Super-telephoto imaging on mobile phones suffers from complex, device-dependent degradations that are poorly captured by conventional handcrafted pipelines. Collecting large-scale aligned training pairs across different phones and diverse scenes, however, is both costly and difficult to scale. We therefore study whether device-specific degradations can be learned from only a small number of physical captures and transferred for training a unified super-resolution (SR) model. Our key idea is to factor degradation learning into two learnable components: one captures the device-specific imaging characteristics under well-resolved conditions, while the other captures the additional information loss and artifacts that arise in super-telephoto imaging. We build a paired benchmark across three phone brands (Huawei, Apple, and vivo) by photographing the same content at two display sizes and zoom settings. The larger displayed image is captured at the phone's maximum optical zoom setting, while the smaller image is photographed directly at the phone's highest native zoom setting through the phone's native imaging pipeline. The benchmark contains 519 training and 60 held-out pairs. We use the benchmark to learn device-specific degradation, then apply it to external images to synthesize thousands of additional pairs to train our UniTele SR model. Relative to the pretrained Qwen and FLUX backbones before adaptation, the full UniTele training recipe substantially improves paired restoration fidelity and perceptual similarity on the held-out benchmark and yields markedly more stable and visually faithful restorations.
Nhan Trong Luu, Duong Trung Luu, Thang Cong Truong
Spiking neural networks (SNNs) promise event-driven, low-power inference, but on vision tasks they still acquire rotation invariance only through data augmentation. Group-equivariant CNNs build such symmetries into the architecture, yet a faithful spiking analogue is missing: a leaky integrate-and-fire (LIF) neuron placed after a $G$-equivariant convolution can break equivariance through its membrane recurrence. We give a necessary and sufficient condition for this not to happen. The layer is equivariant at every time step if and only if the neuron parameters $(β, v_{\mathrm{th}})$ are constant along the joint orbits of the group action on the (group, space) index. We solve it in closed form, and it covers any elementwise stateful neuron. Its sufficiency half specialises a known result for equivariant recurrent networks; its necessity half rules out the heterogeneous per-neuron parameters common in SNNs. The per-channel parameterisation that satisfies it, Group-Coupled LIF (GC-LIF), is combined with $p4$ and $p4m$-equivariant convolutions and a spike-rate group max-pool into SGE-Net. On MNIST, KMNIST, Fashion-MNIST, CIFAR-10 and CIFAR-100, evaluating each checkpoint on the clean test set and all its group-transformed copies, SGE-Net trained without rotations is invariant under $C_4$ and $D_4$ and outperforms a plain SNN trained with rotations in nine of ten dataset-group pairs, whereas non-equivariant baselines lose up to $72$ percentage points in the worst case. Operator-level tests confirm the theory before and after training, and an analytical estimate places SGE-Net at $1.7\times$ the arithmetic energy of a standard CNN and $2.3\times$ below its ANN counterpart at eight time steps.
Eli A. Underwood, Nicolas Rondan, Angel V. Peterchev, J. Matias Di Martino
Aligning a partial, real-world 3D face scan to a canonical representation requires solving two coupled problems at once: estimating the rigid pose that brings the scan into a common frame, and completing the facial shape in regions the scan never observed. Existing pipelines typically decouple these steps, running a generic point-cloud registration front end followed by a separate non-rigid fitting or deformation stage. Building on the GeoTransformer superpoint-matching architecture, we propose a single network that predicts both a rigid pose and a shape-complete point cloud in one forward pass: a cross-attention regressor, conditioned on transformer point features and predicted coarse correspondences, maps a partial scan to per-patch coefficients of a 3D morphable face model learned offline from a corpus of 8,000 identities. The resulting morphed template is then used, in the same forward pass, as the target for fine-grained Sinkhorn matching and rigid pose estimation, supervised purely by geometric alignment to ground-truth morphed geometry, so no hand-labeled coefficients are required. Against six classical and learned non-rigid registration baselines on held-out identities, our method leads the strongest ones by 3-4 times on per-vertex error and baselines that collapse the head onto the visible region by more than an order of magnitude, and is simultaneously the most accurate and fastest method on ground-truth-free scans from two independent real-scan datasets, NoW and NPHM. Code will be made publicly available.
First-derivative edge detectors such as Sobel, Prewitt, Roberts and Scharr measure absolute grey-level differences. We revisit the log-gradient magnitude, a relative-gradient edge operator, and motivate it by analogy with the superderivative, a multiplicative derivative in which the independent variable is perturbed multiplicatively rather than additively. In one and two dimensions the logarithm of the superderivative is a radial logarithmic derivative; its local, origin-free, isotropic analogue is the magnitude of the logarithmic gradient. The resulting operator measures relative rather than absolute change, giving an edge-strength map that is invariant under global intensity rescaling. We present the theory, show how the operator can be implemented in two steps---a classical gradient kernel plus a pointwise division---and compare it with standard first-derivative filters on synthetic images with ground-truth edges and on the official BSDS500 benchmark. We also study a parametric family that interpolates between the absolute gradient and the log-gradient; empirically, the best members are small intermediate exponents, which edge the classical kernels by a small but statistically detectable margin in the per-image metric and, with the same Gaussian smoothing as the single-$σ$ Canny, match it in ODS and lead it in OIS and AP, while the tuned Canny remains ahead in ODS. The single-threshold margin over the kernels does not replicate on the held-out test split, and the fully relative endpoint is never the best member, its ranking depending on the edge-map normalisation. The contribution is the construction and the invariance it buys, not a new state of the art.
Can a Watershed transform, the classical tool of morphological segmentation, be placed at the end of a deep network and trained through? The Power Watershed (PW) is the limit of the seeded Random Walker (RW) as p grows to infinity with edge weights raised to the power p. At every finite p the RW solves a sparse positive-definite linear system and is exactly differentiable with respect to the edge weights and the unaries through one adjoint solve; its multi-label form costs one right-hand side per label, and its limit is the PW. We study this family as the last layer of a segmentation network, the watershed layer (WS layer). A conditioning study measures how the solve, the gradient and the distance to the limit evolve with p, and the implicit gradient is checked against unrolling. On synthetic data with correct edge costs the WS layer transforms a soft prediction into an exact boundary. On nuclei segmentation (PanNuke), a binary WS layer is level with a thresholded U-Net. With the instances as labels, the WS layer turns the U-Net into an instance segmenter whose number of labels is set per image by the markers rather than by the output layout, and the network is trained through that partition: instance F1 score rises from 0.656 to 0.795 over a U-Net with connected components, level with the same network followed by a tuned marker watershed. At the training temperature the partition is a watershed of the learned costs, and the finite temperature lets the solve hedge, discounting weak markers and sharing contested pixels. The hedging puts the WS layer 0.015--0.020 F1 ahead of the baseline when trained on a tenth of the data, at the price of having to deploy the finite-p solve rather than its exact limit at inference. Code and the full experimental record are released.
Language models increasingly create 3D content through code, with practical workflows often relying on editing existing programs rather than regenerating them, for speed and precise control. Such edits require understanding what a program builds and foreseeing how a change reshapes it, yet prior benchmarks mostly evaluate generation or simple edits against a reference. In this paper, we introduce 3DCodeEdit, a benchmark of 2,592 instruction-guided editing tasks over 108 Blender programs that construct objects entirely in code. Its edits range from parameter to structural changes, style redesign, and scene composition, and are judged from renders against human-reviewed checklists. Across 17 models, the average success rate is only 37%, and even the strongest models complete fewer than 60% of structural and style edits. As models become more capable, their failures shift to geometric foresight, where the code runs but builds the wrong shape. We therefore study whether multi-turn feedback can compensate for this missing foresight. Feedback from a separate reviewer model that describes how the renders differ from the request helps most, more than maximum reasoning effort or an agent harness, and can let a 27B model surpass the single-shot rate of a 2.8T-parameter model.
Ball bounces and racket hits are the events from which tennis statistics are derived, and on monocular video the two are easy to confuse. We release a benchmark of 345 bounce and hit annotations at frame level on eight clips that mix broadcast, fixed elevated, phone and court-level viewpoints, with a frozen development and test split, a confusion-aware matcher, clip-level bootstrap intervals and a blind second annotation of three clips. On it we score supervised precise spotters with published broadcast weights, two zero-shot vision-language models, and RallyArb, an arbitration layer over a ball track and player poses that fits no parameter on event labels. The supervised spotters score near-perfectly on broadcast footage and almost nothing on phone and court-level footage, where the pooled bounce F1 of E2E-Spot is 0.130. RallyArb keeps a bounce F1 above 0.8 on the three phone and court-level clips, two of which were used to tune it, and finds 8 of 11 bounces on the held-out one, but it places hits too far from the contact frame to match human timing. The vision-language models detect events on every viewpoint and miss the contact frame by about 0.1 s. Fine-tuned leave-one-clip-out from its broadcast weights, the spotter reaches about 0.8 bounce F1 on two phone and court-level clips and stays at 0.2 on a grazing phone angle. The annotations, the evaluator and the frozen predictions of every scored row are released.
World action models (WAMs) predict the visual consequences of the actions they generate, which makes test-time selection a natural way to improve a fixed policy: imagine the futures of candidate actions and execute the best one. The benefit of selection is bounded by the selectable headroom, how much better the best candidate truly is, but the standard estimate, which executes each candidate once and keeps the best outcome, conflates candidate quality with the randomness of everything that follows. We introduce a denoised counterfactual protocol, based on bitwise-reproducible branching, repeated continuations, held-out oracles, and common random numbers, that measures the headroom a selector can realize, and use its measurements to build a fail-gated selector. Across four WAMs, two benchmarks, over 100,000 branched rollouts, and pre-registered tests, we find that (i) the standard estimate overstates the selectable headroom of native samples: bounds from 17 continuations per candidate place its selectable share between 24% and 66% for three WAMs; (ii) structured action perturbations create up to 16 points of selectable headroom; and (iii) in both WAMs with action-conditioned imagination, imagined futures carry candidate-outcome information with no detectable loss relative to real futures, information that the value head of Cosmos Policy leaves unused. Deviating from the policy chunk only when a readout of the WAM predicts failure, the selector raises closed-loop task success from 76.9% to 81.6% on held-out episodes of training tasks, with the largest gain where the policy fails most often, and from 62.6% to 69.8% on unseen tasks.
Open-Set Test-Time Adaptation (OSTTA) operates on non-stationary target streams where covariate and semantic shifts coexist. Existing TTA methods often obtain adaptation signals by updating model parameters or maintaining mutable memories, which can be costly and vulnerable to unknown samples. We present TALOR, a lightweight, backward-free rectification module that uses the source linear head as a geometric anchor. The head-induced basis decomposes each normalized feature into high-gain principal coordinates and low-gain tail coordinates. TALOR estimates a soft-routed affine correction in the principal subspace, combining a regime-level principal bias with a tail-conditioned slope that captures sample-level tail-to-principal coupling. Using weighted ridge regression, it subtracts this drift estimate from the principal coordinates, reconstructs the feature and feeds it into the source head. On the ImageNet-C benchmark with six csOOD datasets, TALOR outperforms the next-best UniEnt by 2.2% H-score while running 2.5x faster and using only 12% of its GPU memory; as a post-correction plug-in, it further boosts COME to 68.4% H-score with only 2% additional memory overhead.
Yan Wang, Jiangyong Huang, Siyuan Huang, Baoxiong Jia
Vision-language navigation (VLN) requires an embodied agent to follow language instructions through 3D environments. Maintaining persistent episodic memory remains a challenge for VLM-based streaming agents in long-horizon navigation. We present MEMVLN, a streaming episodic memory framework that maintains a bounded bank of compressed perceptual features annotated with timesteps and action-derived relative poses, and produces compact memory tokens via cross-attention. We pair the proposed memory with a frozen pretrained geometry encoder. To learn the memory module, we design a masking-based trajectory summarization task with multi-granularity sub-trajectory decomposition, encouraging route reconstruction exclusively from episodic memory. Under navigation training restricted to the official R2R/RxR trajectories, MEMVLN achieves state-of-the-art performance on most reported metrics on standard VLN-CE benchmarks. Long-horizon analyses further characterize when episodic memory is useful. For physical evaluation, we collect 30 instruction-conditioned episodes in one physical scene, use 24 as demonstrations for post-training, and hold out six for testing. On the held-out episodes, MEMVLN improves success over the VLM baseline from 17% to 40%, and post-training further raises it to 47%.
Vision-language models (VLMs) ground a referring expression in one of two ways. They decode the box as coordinate tokens, which is precise but takes one sequential step per token, or they regress it from a hidden state with a small head, which takes one step but loses accuracy at high IoU. The two fail differently: a head usually finds the right object but places its edges loosely, while coordinate tokens place edges precisely. We therefore use the regression head as a speculative drafter for the model's own coordinate tokens. Locus writes the head's box as coordinate tokens and checks them with the same backbone by Jacobi iteration over a shared image prefix. The fixed point is the model's greedy coordinate-token answer, reached in about 5.60 sequential passes on average instead of 18.46 decoding steps; head and verifier are one set of weights, trained jointly. On the 8 RefCOCO, RefCOCO+ and RefCOCOg splits, verification raises Locus-2B's Acc@0.9 from 54.87 for its own head to 59.13, and the one-pass head is ahead at Acc@0.5 of every published model with at most 3B parameters except Florence-2-L. Against Jag, the strongest open regression-head model, at its own size and training budget, Locus ties at Acc@0.5 and, with typical acceptance, gains +4.52 points at Acc@0.9, with paired significance tests. Exact answers come 1.5x faster than vLLM greedy-decoding the same coordinate tokens with the same weights; a pure regression head remains faster still.
The rapid development of mobile imaging and social platforms has led to explosive growth in user-generated content images, making their systematic and comprehensive perceptual assessment increasingly important. Existing image perception benchmarks are typically task-oriented, curating specific scenarios for quality or aesthetic assessment and applying a uniform perceptual criterion across diverse scenarios. This limits their ability to capture the scene diversity and perceptual complexity of real-world UGC images. To address these limitations, we introduce ScenePercept-Bench, a benchmark designed to reflect the distribution of real-world UGC images, comprising 23,100 images across 21 categories and 105 fine-grained scenes. To account for the influence of scene context on human perception, we design a hierarchical, scene-aware annotation protocol, covering visual quality, aesthetic perception, and scene semantics, with 15 attributes in total. Each image is annotated with an overall score, 15 attribute-level scores, and structured explanatory reasoning. Building on this benchmark, we further introduce Scenario-Conditioned Perceptual Alignment, a training framework that tackles the heterogeneity of perceptual criteria across scenarios. SCPA jointly optimizes absolute score prediction, preference ranking, and cross-scenario calibration, preserving the model general perceptual assessment ability while encouraging it to explore and adapt perceptual criteria across different scenarios. Extensive experiments demonstrate that SCPA achieves state-of-the-art performance in UGC image perception and generalizes effectively to established IQA and IAA benchmarks.
Natasha Ironside, Hongyu Kan, Kristofor Pas, Mohammad Shifat-E-Rabbi, Benjamin Newman, Shinjini Kundu, Kareem El Naamani, Tanvir Rizvi 等 20 位作者
Background The combination of hematoma location, expansion and final volume influence neurological outcome. The aim of this study was to derive and validate a quantitative and interpretable region-based NCCT image feature model for predicting hematoma expansion. Methods The Ethnic/Racial Variations of Intracerebral Hemorrhage(ERICH) study was used for derivation and the Virtual International Trials in Stroke Archive(VISTA) was used for validation. NCCT images at presentation and 24+/-6hours were pre-processed, segmented and categorized into lobar, basal ganglia or thalamic regions. Hematoma expansion was defined by region and according to its effect on clinical outcome. Transport-based morphometry (TBM) produced a unique transport-based representation from each NCCT hematoma image. Linear classifiers applied to transport representations were used to derive a predictive model. The regional TBM model's performance was compared to alternate methods using AUROC analyses. Results The derivation dataset included 574 patients, comprising 162 lobar, 257 basal ganglia and 155 thalamic hematomas. The validation dataset included 196 patients, comprising 31 lobar, 119 basal ganglia and 46 thalamic hematomas. A combined regional model predicted hematoma expansion in the validation dataset with an AUROC of 0.787. The principal image features which varied by location were size, heterogeneity, eccentricity and density distribution towards the hematoma periphery. The regional TBM model outperformed alternative clinician based and machine-learning methods. Conclusions TBM discriminated and visualized regional features of hematoma expansion from presentation NCCT scans. This provides preliminary evidence for regional radiographic features of hematoma expansion. It has the potential to improve hematoma expansion prediction and understanding of hematoma expansion pathophysiology.
Multimodal Emotion Recognition (MER) systems often suffer from missing modalities in real-world scenarios. Existing methods usually generate, align, or distill missing modalities as a whole, overlooking the heterogeneous nature of the information carried by each modality. Such holistic treatment mixes inferable shared semantics with uncertain modality-specific details, yielding unstable representations and degrading robustness. To address this issue, we propose the Primitive Memory Distillation (PriMD) framework. Unlike existing methods, PriMD takes an intra-modal perspective and focuses on how different types of information within a modality differ in recoverability. PriMD first disentangles cross-modal shared semantics from modality-specific representations, and then discretizes the latter into learnable semantic primitives to construct modality-specific memory banks. When modalities are missing, PriMD is a teacher-student framework that the student model uses the shared semantics of available modalities as queries to dynamically retrieve primitives. It compensates for missing modality-specific information within a constrained memory space and aligns with the teacher model. Extensive experiments on IEMOCAP, CMU-MOSI, and CMU-MOSEI demonstrate that PriMD achieves state-of-the-art performance and consistently stronger robustness across a wide range of missing-modality settings, while mitigating the instability caused by holistic feature inference. Our code and project website are available at https://www.jiaqizhang-sengoku.com/Code/PriMD and https://www.jiaqizhang-sengoku.com/PriMD/ , respectively.
Bowen Zeng, Feiyang Ren, Jun Zhang Qwen, Xiaoling Gu Hangzhou Dianzi University, Ke Chen Zhejiang University, Lidan Shou Zhejiang University, Huan Li Zhejiang University
Multimodal Large Language Models (MLLMs) have advanced unified reasoning over text, images, and videos, but their inference is hindered by the rapid growth of key–value (KV) caches. Each visual input expands into thousands of tokens, causing caches to scale linearly with context length and remain resident in GPU memory throughout decoding, which leads to prohibitive memory overhead and latency even on high-end GPUs. A common solution is to compress caches under a fixed allocated budget at different granularities: token-level uniformly discards less important tokens, layer-level varies retention across layers, and head-level redistributes budgets across heads. Yet these approaches stop at allocation and overlook the heterogeneous behaviors of attention heads that require distinct compression strategies. We propose HybridKV, a hybrid KV cache compression framework that integrates complementary strategies in three stages: heads are first classified into static or dynamic types using text-centric attention; then a top-down budget allocation scheme hierarchically assigns KV budgets; finally, static heads are compressed by text-prior pruning and dynamic heads by chunk-wise retrieval. Experiments on 11 multimodal benchmarks with Qwen2.5-VL-7B show that HybridKV reduces KV cache memory by up to 7.9\times and achieves 1.52\times faster decoding, with almost no performance drop or even higher relative to the full-cache MLLM.
尚无 AI 解读，展示原文摘要
cs.CL · cs.CV
Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表
Herzog, Henry, Bastani, Favyen, Zhang, Yawen, Tseng, Gabriel, Redmon, Joseph, Sablon, Hadrien, Park, Ryan, Morrison, Jacob 等 26 位作者
Earth observation data presents a unique challenge: it is spatial like images, sequential like video or text, and highly multimodal. We present Helios: a multimodal, spatio-temporal foundation model that employs a novel self-supervised learning formulation, masking strategy, and loss all designed for the Earth observation domain. Helios achieves state-of-the-art performance compared to 12 other foundation models across a variety of research benchmarks and real-world tasks from external partners. When evaluating embeddings Helios achieves the best performance on 15 out of 24 tasks, and with full fine-tuning it is the best on 19 of 29 tasks. We deploy Helios as the backbone of an end-to-end platform for data collection, labeling, training, and inference of Earth observation models. The Helios platform puts frontier foundation models and powerful data management tools into the hands of non-profits and NGOs working to solve the world's biggest problems. Helios source code, training data, and pre-trained weights are available at REDACTED.
尚无 AI 解读，展示原文摘要
cs.CV
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition · 2026 · 正式发表
The performance of Vision-Language Transformers drops sharply when an input modality (e.g., image) is missing, because the model is forced to make predictions using incomplete information. Existing missing-aware prompt methods help reduce this degradation, but they still rely on conventional prediction heads (e.g., a Fully-Connected layer) that compute class scores in the same way regardless of which modality is present or absent. We introduce Decoupled Prototype Learning (DPL), a new prediction head architecture that explicitly adjusts its decision process to the observed input modalities. For each class, DPL selects a set of prototypes specific to the current missing-modality cases (image-missing, text-missing, or mixed-missing). Each prototype is then decomposed into image-specific and text-specific components, enabling the head to make decisions that depend on the information actually present. This adaptive design allows DPL to handle inputs with missing modalities more effectively while remaining fully compatible with existing prompt-based frameworks. Extensive experiments on MM-IMDb, UPMC Food-101, and Hateful Memes demonstrate that DPL outperforms state-of-the-art approaches across all widely used multimodal image-text datasets and various missing cases.
尚无 AI 解读，展示原文摘要
cs.CV
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition · 2026 · 正式发表
Federated learning (FL) enables a central server to collaboratively train a global model with multiple clients while preserving data privacy. However, the distributed nature of FL makes the paradigm vulnerable to backdoor attacks, as proved by numerous recent studies. Although existing studies improve the effectiveness of backdoor attacks through optimized triggers, they have two limitations: (1) they ignore the heterogeneous contribution of individual model layers to the success of a backdoor; (2) they induce conspicuous differences between backdoor and clean models in the early stages of poisoning. The limitations cause backdoor models to exhibit significant discrepancies from clean models, making them easily detectable. To fill these gaps, we propose LaySelFL, a novel layer-selective method to eliminate distance differences induced by the backdoor to conceal attacks in FL. Our central insight is that different layers contribute unequally to backdoor attacks, by localizing poisoning to layers that are most sensitive to backdoor objectives, an attacker can reduce the model differences substantially between the backdoor and clean models. Concretely, LaySelFL identifies sensitive layers via both dynamic and static evaluations of parameter differences between backdoor and benign models, and then applies a targeted training protocol and a regularized loss that constrains differences from the global model in each round. Finally, LaySelFL performs clipping on non-poisoning layers to further mask residual differences introduced by the attack. This strategy yields a more covert and resilient backdoor attack. Extensive experiments show that LaySelFL increases the effectiveness of attacks by 25% and reduces the effectiveness of defense methods to 4%.
尚无 AI 解读，展示原文摘要
cs.CV
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition · 2026 · 正式发表
You, Xiaoxing, Huang, Qiang, Li, Lingyu, Chang, Xiaojun, Yu, Jun
Multimodal Summarization (MMS) aims to generate concise textual summaries by understanding and integrating information across videos, transcripts, and images. However, existing approaches still suffer from three main challenges: (1) reliance on domain-specific supervision, (2) implicit fusion with weak cross-modal grounding, and (3) flat temporal modeling without event transitions. To address these issues, we introduce **CoE**, a training-free MMS framework that performs structured reasoning through a **Chain-of-Events** guided by a Hierarchical Event Graph (HEG). The HEG encodes textual semantics into an explicit event hierarchy that scaffolds cross-modal grounding and temporal reasoning. Guided by this structure, **CoE** localizes key visual cues, models event evolution and causal transitions, and refines outputs via lightweight style adaptation for domain alignment. Extensive experiments on eight diverse datasets demonstrate that **CoE** consistently outperforms state-of-the-art video CoT baselines, achieving average gains of **+3.04 ROUGE**, **+9.51 CIDEr**, and **+1.88 BERTScore**, highlighting its robustness, interpretability, and cross-domain generalization. Our code is available at https://github.com/youxiaoxing/CoE.
尚无 AI 解读，展示原文摘要
cs.CV
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition · 2026 · 正式发表
We present PercHead, a model for single-image 3D head reconstruction and disentangled 3D editing - two tasks that are inherently challenging due to ambiguity in plausible explanations for the same input. At the heart of our approach lies our novel perceptual loss based on DINOv2 and SAM 2.1. Unlike widely-adopted low-level losses like LPIPS, SSIM or L1, we rely on deep visual understanding of images and the resulting generalized supervision signals. We show that our new loss can be a drop-in replacement for standard losses and used to improve visual quality in high-frequency areas. We base our model architecture on Vision Transformers (ViTs), allowing us to decouple the 3D representation from the 2D input. We train our method on multi-view images for view-consistency and in-the-wild images for strong transferability to new environments. Our model achieves state-of-the-art performance in novel-view synthesis and, furthermore, exhibits exceptional robustness to extreme viewing angles. We also extend our base model to disentangled 3D editing by swapping the encoder and fine-tuning the network. A segmentation map controls geometry and either a text prompt or a reference image specifies appearance. We highlight the intuitive and powerful 3D editing capabilities through an interactive GUI.
尚无 AI 解读，展示原文摘要
cs.CV
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition · 2026 · 正式发表
Diffusion- and flow-based models usually allocate compute uniformly across space, updating all patches with the same timestep and number of function evaluations. While convenient, this ignores the heterogeneity of natural images: some regions are easy to denoise, whereas others benefit from more refinement or additional context. Motivated by this, we explore patch-level noise scales for image synthesis. We find that naively varying timesteps across image tokens performs poorly, as it exposes the model to overly informative training states that do not occur at inference. We therefore introduce a timestep sampler that explicitly controls the maximum patch-level information available during training, and show that moving from global to patch-level timesteps already improves image generation over standard baselines. By further augmenting the model with a lightweight per-patch difficulty head, we enable adaptive samplers that allocate compute dynamically where it is most needed. Combined with noise levels varying over both space and diffusion time, this yields Patch Forcing (PF), a framework that advances easier regions earlier so they can provide context for harder ones. PF achieves superior results on class-conditional ImageNet, remains orthogonal to representation alignment and guidance methods, and scales to text-to-image synthesis. Our results suggest that patch-level denoising schedules provide a promising foundation for adaptive image generation.
尚无 AI 解读，展示原文摘要
cs.CV
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition · 2026 · 正式发表
We propose OMG-Avatar, a novel One-shot method that leverages a Multi-LOD (Level-of-Detail) Gaussian representation for animatable 3D head reconstruction from a single image in 0.2s. Our method enables LOD head avatar modeling using a unified model that accommodates diverse hardware capabilities and inference speed requirements. To capture both global and local facial characteristics, we employ a transformer-based architecture for global feature extraction and projection-based sampling for local feature acquisition. These features are effectively fused under the guidance of a depth buffer, ensuring occlusion plausibility. We further introduce a coarse-to-fine learning paradigm to support Level-of-Detail functionality and enhance the perception of hierarchical details. To address the limitations of 3DMMs in modeling non-head regions such as the shoulders, we introduce a multi-region decomposition scheme in which the head and shoulders are predicted separately and then integrated through cross-region combination. Extensive experiments demonstrate that OMG-Avatar outperforms state-of-the-art methods in reconstruction quality, reenactment performance, and computational efficiency.
尚无 AI 解读，展示原文摘要
cs.CV
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition · 2026 · 正式发表
Federated Graph Learning (FGL) has emerged as a principled framework for decentralized training of Graph Neural Networks (GNNs) while preserving data privacy. In subgraph-FL scenarios, however, structural noise arising from data collection and storage can damage the GNN message-passing scheme of clients, leading to conflicts in collaboration. Existing approaches exhibit two critical limitations: 1) Globally, they fail to identify corrupted clients, causing destructive knowledge inconsistencies. 2) Locally, the global GNN performs poorly on these clients due to structural noise, limiting their ability to benefit from federated collaboration. To address these challenges, we propose FedSDR, a spectra-based FGL framework against high-structural-noise scenarios. Specifically, Structural Noise-Aware Aggregation (SNAA) introduces a structural fidelity evaluation metric to detect corrupted clients and reduce their contributions, thereby mitigating the impact of noise on the global GNN. Furthermore, Robust Local Structure Reconstruction (RLSR) leverages the knowledge from the healthy global model to repair locally corrupted graph structures. Extensive experiments demonstrate that FedSDR outperforms state-of-the-art methods across various scenarios under structural noise. The code is available at https://github.com/Subtleazure/FedSDR.
尚无 AI 解读，展示原文摘要
cs.CV
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition · 2026 · 正式发表
Realistic digital avatars require expressive and dynamic hair motion; however, most existing head avatar methods assume rigid hair movement. These methods often fail to disentangle hair from the head, representing it as a simple outer shell and failing to capture its natural volumetric behavior. In this paper, we address these limitations by introducing PhysHead, a hybrid representation for animatable head avatars with realistic hair dynamics learned from multiview video. At the core is a 3D Gaussian-based layered representation of the head. Our approach combines a 3D parametric mesh for the head with strand-based hair, which can be directly simulated using physics engines. For the appearance model, we employ Gaussian primitives attached to both the head mesh and hair segments. This representation enables the creation of photorealistic head avatars with dynamic hair behavior, such as wind-blown motion, overcoming the constraints of rigid hair in existing methods. However, these animation capabilities also require new training schemes. In particular, we propose the use of VLM-based models to generate appearance of regions that are occluded in the dynamic training sequences. In quantitative and qualitative studies, we demonstrate the capabilities of the proposed model and compare it with existing baselines. We show that our method can synthesize physically plausible hair motion besides expression and camera control. For additional results and code, please refer to https://physhead.github.io
尚无 AI 解读，展示原文摘要
cs.CV
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition · 2026 · 正式发表
Computer-Aided Design (CAD) is essential in industrial design, but the complexity of traditional CAD modeling and workflows presents significant challenges for automating the generation of high-precision, editable CAD models. Existing methods, such as 3D reconstruction from sketches, often produce non-editable, approximate models that fall short of meeting the stringent requirements for precision and editability in industrial design. Moreover, the reliance on text or image-based inputs often requires significant manual annotation, limiting their scalability and applicability in industrial settings. To overcome these challenges, we propose the Heterogeneous Collaborative Multi-Expert Reinforcement Learning (CME-CAD) paradigm, a novel training paradigm for CAD code generation. Our approach integrates the complementary strengths of these models, facilitating collaborative learning and improving the model's ability to generate accurate, constraint-compatible, and fully editable CAD models. We introduce a two-stage training process: Multi-Expert Fine-Tuning (MEFT), and Multi-Expert Reinforcement Learning (MERL). Additionally, we present CADExpert, an open-source benchmark consisting of 17,299 instances, including orthographic projections with precise dimension annotations, expert-generated Chain-of-Thought (CoT) processes, executable CADQuery code, and rendered 3D models. The CADExpert dataset is publicly available at https://modelscope.cn/datasets/zhuofanChen/CADExpert.
尚无 AI 解读，展示原文摘要
cs.CV
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition · 2026 · 正式发表
Recent works on vision-language-action (VLA) models have made great progress in exploring action tokenizers that convert continuous control signals into discrete tokens to align with LLM/VLM training paradigms.These approaches typically train a single tokenizer over entire manipulation trajectories, which often comprise multiple distinct skills and thus pose a challenging optimization trade-off.To address this issue, we introduce MoEActok, a novel action tokenizer that employs a mixture-of-experts (MoE) quantizer to produce skill-aware discrete representations for VLA models. MoEActok utilizes a clustering-driven MoE VQ-VAE mechanism in which each expert specializes in a particular skill.The key components are: (a) an action-skill decoupling strategy that uses k-means clustering to group action chunks, aligning clusters having similar skills; (b) a skill-aware training paradigm that augments VLA models with skill-conditioned context, improving skill grounding; and (c) an adapter that projects shared encoder representations into skill-specific latent spaces for specialized quantization, and subsequently harmonize the heterogeneous quantized representations back into a unified space for coherent reconstruction by the shared decoder.We evaluate MoEActok-based VLA models against multiple prior action tokenizer baselines in the RoboTwin and Simpler-Env simulators, and further assess zero-shot transfer on three real-world tasks. Across both simulated and real-world settings, MoEActok-based VLA substantially outperforms existing discrete tokenization methods.
尚无 AI 解读，展示原文摘要
cs.CV · cs.CL
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition · 2026 · 正式发表
Wang, Weitian, Meiner, Lukas, Shubham, Rai, De La Parra, Cecilia, Kumar, Akash
The Visual Geometry Grounded Transformer (VGGT) marks a significant leap forward in 3D scene reconstruction, as it is the first model that directly infers all key 3D attributes (camera poses, depths, and dense geometry) jointly in one pass. However, this joint inference mechanism requires global attention layers that perform all-to-all attention computation on tokens from all views. For reconstruction of large scenes with long-sequence inputs, this causes a significant latency bottleneck. In this paper, we propose head-wise temporal merging (HTTM), a training-free 3D token merging method for accelerating VGGT. Existing merging techniques merge tokens uniformly across different attention heads, resulting in identical tokens in the layers' output, which hinders the model's representational ability. HTTM tackles this problem by merging tokens in multi-head granularity, which preserves the uniqueness of feature tokens after head concatenation. Additionally, this enables HTTM to leverage the spatial locality and temporal correspondence observed at the head level to achieve higher merging ratios with lower merging costs compared to existing methods. Thus, HTTM achieves up to 7x acceleration with negligible performance drops in a GPU-based inference.
尚无 AI 解读，展示原文摘要
cs.CV
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition · 2026 · 正式发表
Kim, Minyoung E., Yun, Dae Hee, Patel, Aditi V., Hon, Madeline, Guan, Webster, Lee, Taegeon, Nguyen, Brian
Unprecedented visual details of biological structures are being revealed by subcellular-resolution whole-brain 3D microscopy data, enabled by recent advances in intact tissue processing and light-sheet fluorescence microscopy (LSFM). These volumetric data offer rich morphological and spatial cellular information, however, the lack of scalable data processing and analysis methods tailored to these petabyte-scale data poses a substantial challenge for accurate interpretation. Further, existing models for visual tasks such as object detection and classification struggle to generalize to this type of data. To accelerate the development of suitable methods and foundational models, we present CANVAS, a comprehensive set of high-resolution whole mouse brain LSFM benchmark data, encompassing six neuronal and immune cell-type markers, along with cell annotations and a leaderboard. We also demonstrate challenges in generalization of baseline models built on existing architectures, especially due to the heterogeneity in cellular morphology across phenotypes and anatomical locations in the brain. To the best of our knowledge, CANVAS is the first and largest LSFM benchmark that captures intact mouse brain tissue at subcellular level, and includes extensive annotations of cells throughout the brain.
尚无 AI 解读，展示原文摘要
cs.CV
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition · 2026 · 正式发表
Xu, Mengting, Gu, Shi, Lin, Peng, Ma, De, Tang, Huajin, Zheng, Qian, Pan, Gang
As the third generation of neural networks, Spiking Neural Networks (SNNs) have demonstrated remarkable potential across diverse applications owing to their unique temporal dynamics. In recent years, analyzing the robustness of SNNs from a temporal perspective has become an emerging research focus. However, most existing works examine only the overall temporal behavior of SNNs, typically applying adversarial attacks that rely on time-averaged gradients. In this study, we revisit SNN robustness through the lens of temporal granularity, emphasizing the distinct behaviors that occur at individual time steps. We first introduce a Temporal Granularity Attack (TG-Attack), which selectively perturbs gradients at specific time steps. This approach enables a finer-grained evaluation of SNN robustness across time and demonstrates higher attack success rates than traditional gradient-averaging methods. Furthermore, we theoretically show that the robustness of SNNs at a given time step is determined by the Hessian of the input-output gradient at that step, which we define as Temporal Sensitivity (TS). By calculating the Temporal Sensitivity Value (TSV) for each time step, robustness can be effectively estimated without generating adversarial examples. Finally, we propose a Temporal Granularity Regularization (TG-Reg) term that constrains the TSV across all time steps, thereby improving the model's overall robustness. Experimental evaluations confirm that our framework consistently outperforms existing state-of-the-art methods.
尚无 AI 解读，展示原文摘要
cs.CV
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition · 2026 · 正式发表