Vision Encoders

Typical video object-centric learning (VOCL) approaches employ slot-based frameworks that rely on reconstruction-driven encoder-decoder architectures, where learning is mediated by two spatial maps: attention maps from the encoder and object maps from the decoder. As these two distinct maps exhibit different properties, a recent dense alignment strategy attempted to reconcile this discrepancy by enforcing agreement across all spatio-temporal patches via contrastive learning. However, this indiscriminate alignment inadvertently propagates the inherent weaknesses of each module, such as noisy encoder predictions and blurred decoder boundaries. Moreover, computing dense similarities across all pairs incurs a computational cost quadratic in the total number of spatio-temporal patches, severely limiting scalability. Motivated by this, we propose Selective Synergistic Learning (SSync). Instead of exhaustive patch-to-patch alignment, SSync prevents error propagation by selectively distilling only the most reliable cues: leveraging the encoder strictly for boundary refinement and the decoder for interior denoising. This is realized via a pseudo-labeling with linear complexity, eliminating the need for quadratic spatial comparisons. Also, to prevent the reinforcement of architectural biases like slot redundancy, we introduce a transitive pseudo-label merging that consolidates overlapping slots based on spatio-temporal activation consistency. Extensive studies demonstrate that SSync improves decomposition quality and serves as a versatile, plug-and-play module while also exhibiting exceptional robustness to slot configurations. Code is available at github.com/wjun0830/SSync.

Vision encoders for retrieval are typically trained with class-label supervision: each training pair reduces to a scalar that uniformly pushes the embedding apart or pulls it together, as if every visual attribute either differed or matched. A multimodal large language model (MLLM), shown the same pair, can articulate those attributes and use them to predict whether the images share a class. We propose \textbf{SAGA}, a framework that turns this language-grounded, attribute-aware perception into a training signal for the encoder itself. Specifically, we use Group Relative Policy Optimization (GRPO) to reward the MLLM for correct predictions on the vision encoder's tokens. Since correct predictions require those tokens to expose the specific attributes that differ or match between the pair, the gradient pushes the encoder to encode them, replacing the uniform pair-level scalar with attribute-resolved supervision. An auxiliary attention-distillation loss anchors the encoder's embedding to tokens the MLLM attended to, and a standard metric-learning loss shapes the embedding geometry for nearest-neighbour retrieval. The MLLM is frozen throughout and discarded at inference, matching the deployment cost of a metric-learning baseline. SAGA improves Recall@1 by 3 to 6 points over state-of-the-art baselines on CUB-200-2011, Cars-196, FGVC-Aircraft, and iNaturalist Aves on zero-shot image retrieval.

Visualization-based malware detection maps raw binary bytes to grayscale images and applies learned visual classifiers, providing an evasion-resistant and disassembly-free alternative to conventional analysis pipelines. However, executable packing remains a critical failure mode: packed binaries produce high-entropy images that obscure the structural patterns these models rely on. Because packing is also prevalent in benign software (e.g., for compression or copy protection), packing state alone is not a reliable indicator of maliciousness, and existing approaches do not address this challenge within a unified supervised framework. We present ViPER, a Vision-based Packing-Aware Encoder for Robust malware detection. ViPER builds on a LoRA-adapted ViT-B/14 backbone with a dual-head architecture that jointly learns malware classification and packing detection. A packing-aware gating mechanism conditions malware predictions on the inferred packing state, enabling distinct decision boundaries for packed and unpacked inputs. To address packing label skew during training, we employ frequency-weighted losses with stratified sampling over joint class-packing strata. Evaluated on 200,000 Windows PE byteplot images, ViPER achieves a balanced accuracy of 0.8521, ROC-AUC of 0.9260, and AUPR of 0.9279, outperforming representative state-of-the-art baselines across all primary metrics, while attaining a packing detection AUC of 0.9949.

A vision encoder compresses image pixels into semantic embeddings, implicitly acting as a privacy boundary by preserving semantic content while attenuating pixel-local detail required for exact text recovery. Encoder-free vision-language models (VLMs) remove this boundary by routing image patches directly into the language-model token stream, thereby exposing an architectural privacy attack surface: intermediate visual tokens become a pre-output side channel. Under a token-access adversary, decoders invert visual-token streams from two encoder-free VLMs, Gemma4 and Fuyu, recovering recognizable image structure and readable held-out access codes, whereas matched encoder-based controls localize target regions but recover no exact strings. Within-model ablations show that the operative factor is spatial sampling fidelity of the visual-token grid, especially character-direction sampling density, rather than token or value count. The leakage is not limited to exported tokens: Gemma4 layer-0 key-value cache tensors are directly invertible, placing the side channel within KV caches commonly persisted by production serving stacks for decoding efficiency. The attack survives clutter, realistic document degradation, and zero-shot transfer to public document images, and it resists value-level defenses such as additive noise and quantization. Effective mitigation must therefore reduce spatial sampling, making removal of the vision encoder a first-class privacy decision in VLM deployment.

Vision-language models (VLMs) pretrained on large-scale image-text pairs demonstrate strong image-level understanding, but are primarily optimized for global alignment and do not explicitly encode fine-grained anatomical structure, limiting their suitability for spatially precise tasks such as segmentation. We introduce CheXanatomy, a framework that integrates explicit anatomical knowledge into a pretrained VLM through autoregressive token-space supervision. Instead of adding task-specific decoder heads, the model is trained to generate anatomical segmentation masks via next-token prediction. To enable scalable supervision, we synthesize realistic chest radiographs from CT volumes and forward-project CT segmentation labels to obtain anatomically consistent 2D masks. We evaluate the approach on synthetic and real chest radiographs against a U-Net baseline, including ablations on model scale, input resolution, and vision encoder fine-tuning. Autoregressive anatomical supervision achieves performance comparable to specialized convolutional models in-distribution and demonstrates improved geometric robustness under domain shift to real CXR data. In addition, anatomy-pretrained models exhibit improved sample efficiency when adapting to novel localization tasks under limited supervision. Larger models and higher input image resolution improve performance, while vision encoder fine-tuning has limited effect. These results show that embedding anatomical structure directly into the generative objective promotes spatially grounded representations and supports anatomy-aware medical vision-language modeling.

Different vision neural networks -- trained to classify, contrast, reconstruct, or match images to text -- should have correspondingly different internal representations. We report that they do not. After training, the top sixteen principal directions of variation inside thirteen modern vision encoders converge to the same sixteen-dimensional geometric object. We call this the cross-architecture substrate and study it with PCA, centred kernel alignment (CKA), and Pang 2026 calibration. The substrate transports across four visual domains (natural photographs, medical CT, satellite, microscopy) at median Procrustes-CKA 0.679, and across eight domains (adding sketches, depth, thermal infrared, astronomy) at 0.604, every pair >0.40. It survives Pang calibration globally (7.4x disc-vs-MAE separation, n=13,394) and locally (4.82-5.30, p<10^{-44}). It is not pixel statistics (0.263), not Gabor features (0.31), not a random projection (0.041), and emerges in the first 10% of training while accuracy keeps climbing. We deliver four applications: a label-free transferability filter beating LogME (3x faster, +0.15 Kendall-tau); a four-way domain detector (99.6% accuracy); a frozen low-shot probe (16 dims beat 768-dim DINOv2 by 3.78pp at N=50 labels per class); and a teacher-free distillation auxiliary matching trained-teacher KD on 33 pairs (7.56pp peak gain at 10% label fraction). The substrate does not cross modalities, does not help cross-paradigm distillation, and does not predict transfer quality (rho=0.08 against transfer accuracy).

Recent breakthroughs in self-supervised learning (SSL), such as the Latent-Euclidean Joint-Embedding Predictive Architecture (LeJEPA), alongside successes in integrating visual encoders with language models, have driven the demand for adaptable, high-capacity vision encoders in Computed Tomography (CT). In this work, we explore 2D slice-based architectures as a flexible alternative to native 3D models for processing volumetric CT data. Using the CT-RATE dataset, we trained DALE-CT (Depth-Aware Latent-Euclidean Computed Tomography), a 2D model family built entirely from scratch using LeJEPA, and compared its performance against a continually pre-trained DINOv2 baseline. To enhance representation quality, we developed a novel 3D depth-aware pre-training strategy anchored by dense auxiliary supervision from both automated anatomical masks and human-annotated abnormalities. Under linear probe evaluation with Multiple Instance Learning (MIL) for multi-abnormality detection, the frozen backbone of this dual-supervised model (DALE-CT-2S) achieves a Macro AUROC of 0.833. This performance demonstrates near-parity with state-of-the-art 3D vision-language models, achieved entirely from scratch with significantly less data and no textual supervision. To ensure reproducibility, all training code, evaluation scripts, and model weights have been made publicly available.

We present T-SAR-JEPA, a self-supervised framework for temporal anomaly detection in SAR amplitude stacks via latent prediction. A ViT-Base/16 encoder from SAR-JEPA is domain-adapted on 39,300 Capella patches using local masked reconstruction with gradient feature prediction. A temporal transformer with sinusoidal time encoding forecasts future latent states from K=7 acquisitions, with progressive unfreezing substantially reducing validation loss. The model operates on amplitude alone; InSAR coherence serves exclusively as independent pseudo-ground-truth. On the DFC 2026 dataset (300 time-series, three AOIs), T-SAR-JEPA achieves ROC-AUC of 77.0% on the Hawaii eruption window, outperforming RX, PaDiM, Linear AR, and LSTM baselines (~50%). Spatial coherence of 99.9% (p < 0.001, permutation test) confirms structured detections. Code: https://github.com/TerraLatent/t-sar-jepa

Vision-language models (VLMs) are increasingly used in multi-image, multi-turn agentic settings where decisions depend on visual changes. However, in existing open-weight VLMs, visual comparisons happen only inside the language model, while the visual encoder itself remains stateless: each image is encoded independently, without access to the prior visual context. As a result, small but task-critical changes may be attenuated before the language model has a chance to compare them, especially when those changes do not affect the high-level semantics of the scene. We introduce a Stateful Visual Encoder, which conditions each visual representation on prior visual features. Under supervised finetuning, VLMs equipped with stateful encoders achieve consistent improvements on controlled tasks involving cross-image spatial aggregation, multi-object visual differencing, and visual trajectory behavior cloning. These improvements are consistent across input resolutions, language model sizes, and VLM backbones. Finally, we validate our model on real-world tasks, including longitudinal radiology, fine-grained image comparison, and remote sensing, where stateful encoders consistently improve generalist VLM baselines and can match or surpass specialized models in selected domains. Project page: https://statefulvisualencoders.github.io/

Vision-language models map visual features into a shared embedding space through learned projection layers, yet it remains unclear how these transformations alter the structure of visual information. This study examines changes in representation through spatial-frequency accessibility, measured by the linear recoverability of band-limited Fourier energy from model representations. To isolate effects beyond dimensionality reduction, we introduce Residual Spectral Loss (RSL), which evaluates changes relative to a dimension-matched random projection baseline. To reduce confounding effects from optimization, the analysis uses pretrained models with all parameters frozen. The experimental results show consistent frequency-dependent changes in accessibility across CLIP and DINOv2 on ImageNet and MS-COCO datasets. Spectral accessibility follows a non-monotonic trajectory across depth, peaking at intermediate layers before decreasing toward the output representation. The final transformation differs across architectures: CLIP's learned projection is spectrally neutral, with changes explained by compression, whereas DINOv2's [CLS] pooling induces a structured loss across the spectrum. These findings identify intermediate layers and pooling mechanisms as primary drivers of spectral transformation in modern vision encoders.

The joint optimization of image-based (I2I) and text-based (T2I) person re-identification (ReID) is hindered by modality discrepancies and conflicting training objectives, leading to suboptimal shared representations. While I2I ReID focuses on identity-level invariance across images of the same person, T2I ReID is driven by instance-specific textual descriptions tied to unique visual traits. This paper explores the fundamental difference between two ReID tasks and their optimization processes for effective training. Since I2I and T2I ReID are often studied separately, the loss functions optimized for one retrieval setting may negatively affect the representation quality required by the other. Motivated by these findings, we propose a decoupled two-stage training pipeline for learning a shared representation across image and text modalities. The pipeline is based on a single vision encoder that supports both I2I and T2I retrieval while avoiding cross-task interference during training. We provide extensive experiments across multiple configurations, varying domain mixing procedures, learning strategies, and task objectives. We observed that I2I ReID pre-training positively impacts the generalization ability to T2I data. Besides, we find that incorporating textual supervision during the vision encoder training stage enhances both I2I and T2I performance. We believe our insights provide a meaningful step toward unified ReID systems and cross-modal retrieval overall.

Large vision-language models (LVLMs) achieve strong performance on image and video understanding tasks, but their inference efficiency is constrained by the large number of visual tokens produced by vision encoders. Most existing visual token compression methods estimate token importance from attention scores or representation properties at specific layers, overlooking how visual tokens evolve across the vision encoder. Such layer-specific criteria may provide incomplete importance estimates and limit performance preservation after compression. To address this issue, we analyze layer-wise visual token evolution directions and observe that tokens form multiple group evolution directions across vision-encoder layers. Our analysis further shows that informative tokens tend to exhibit persistent deviations from common group evolution directions. Based on this observation, we propose EvoCut, a training-free and attention-free visual token compression method that estimates token importance from multi-layer evolution deviation. Experimental results show that EvoCut can retain only 11.1% of the visual tokens on LLaVA-1.5-7B while preserving 94.4% of the average performance, demonstrating its effectiveness in balancing efficiency and accuracy.

Vision foundation models are bottlenecked by the quadratic cost of self-attention, which limits usable resolution and increases the cost of large-scale pretraining. Subquadratic alternatives such as linear attention and state-space models reduce this cost, but often serialize images into 1D token streams and weaken the 2D spatial structure important for vision. Generalized Spatial Propagation Networks (GSPN) instead propagate context directly on the 2D grid through line-scan recurrences, achieving near-linear complexity without positional embeddings, but have seen little use as foundation-scale encoders. We present C-GSPN, a foundation-scale vision encoder based on 2D spatial propagation. C-GSPN makes the operator practical through three improvements: (1) a fast GSPN CUDA kernel that fuses per-step launches into a single warp-specialized implementation with shared-memory tiling, coalesced access, and a compact multi-channel propagation, reaching over 90% of peak memory bandwidth and running up to 40--52x faster than the original GSPN implementation; (2) a compressed latent-space propagation block with fused normalization, which turns kernel-level speed into block- and model-level efficiency; and (3) a two-stage cross-operator distillation recipe that trains the new architecture from an attention teacher without the cost of from-scratch foundation-scale training. Distilled with 600M image-text pairs, C-GSPN matches an isomorphic ViT baseline with 15% fewer parameters, improves ADE20K segmentation by +2.1%, transfers to high resolution with a fraction of the data needed from scratch, and delivers a 4x end-to-end block speedup at 2K with single-pass, tiling-free inference.

Recent diffusion transformers have demonstrated strong image synthesis capabilities but remain inefficient to train due to weak alignment between generative and discriminative representations. While representation alignment frameworks such as REPA improve convergence by aligning noisy denoising features with pretrained visual encoders, their externally supervised alignment loss is static and lacks adaptivity during training and inference. Existing methods rely on fixed cosine alignment or contrastive objectives, which cannot dynamically balance representation consistency and generation quality, resulting in limited discriminative benefit and failing to optimize alignment in a task-adaptive manner. To address this, we propose VRPO, a reinforcement-based optimization strategy that replaces REPA's static alignment loss with a generative representation policy optimization objective. Instead of enforcing a fixed similarity constraint, VRPO treats representation alignment as a reward-guided process: the model receives adaptive rewards based on generation fidelity, perceptual quality, and semantic coherence between the diffusion features and pretrained visual embeddings. This formulation enables the generator to continuously refine its internal representations toward semantically meaningful directions while improving image quality. Our VRPO-driven training seamlessly integrates into diffusion transformers, introducing negligible computation cost and preserving full compatibility with SiT and DiT architectures. Extensive experiments on ImageNet-256x256 demonstrate that our VRPO-Alignment substantially enhances both convergence and fidelity, achieving up to +1.8 FID improvement and 2.3x faster training compared to REPA under identical compute budgets.

Learning generalizable vision-language-action (VLA) models from large-scale human videos is promising but challenging due to cross-embodiment discrepancies in both visual observations and executable actions. While latent action models reduce the action execution gap by learning action abstractions, they still rely on visual features. Thus, misaligned human and robot visual representations can lead to inconsistencies in policy inputs and induce domain-dependent latent actions, hindering effective co-training with human videos. To address this, we propose HARP, a human-robot aligned representation learning framework for more effective VLA pretraining from human videos. Specifically, HARP uses limited paired human-robot demonstrations as cross-embodiment bridges and abundant unpaired human and robot videos as a scalable dynamics supervision data source. It trains a robot-adapted visual encoder and a latent action model with manipulation-centric auxiliary cues and a source-relative pair-discriminative alignment loss, which adapts robot representations toward human semantics while preserving pair-level discrimination. The learned aligned vision encoder and latent action model provide a unified vision and action representation for VLA-style policy learning, where human and robot videos provide vision-language-to-latent-action supervision and a lightweight robot action head grounds latent actions into executable commands. Experiments on feature visualization, simulation, and realworld manipulation show improved human-robot alignment and downstream policy performance, achieving 4.481 average length on CALVIN ABC$\rightarrow$D and a 7.1% realworld success rate gain over the strongest baseline.

Video large language models (Video-LLMs) have demonstrated strong capabilities in video understanding tasks. However, their practical deployment is still hindered by the inefficiency introduced by processing massive amounts of visual tokens. Although recent approaches achieve extremely low token retention ratios while maintaining accuracy comparable to full-token baselines, most of them perform compression only at the late stage of prefilling, leaving the efficiency of the vision encoder unoptimized. In this paper, we first show that vision encoding contributes a large portion to the time-to-first-token (TTFT). Therefore, instead of compressing visual tokens only after the vision encoder, performing compression inside the encoder still leaves substantial room for exploration. Based on this insight, we propose EarlyTom, a training-free token compression framework that performs early-stage visual token compression inside the vision encoder, enabling significantly better TTFT reduction and higher throughput. In addition, we introduce a decoupled spatial token selection strategy that improves the overall compression effectiveness. EarlyTom reduces TTFT by up to 2.65x and FLOPs by up to 61% on a single NVIDIA A100 GPU for the LLaVA-OneVision-7B model, while maintaining accuracy comparable to the full-token baseline. These improvements substantially enhance the practicality of deploying Video-LLMs in real-world production scenarios.

Vision-Language-Action (VLA) models have emerged as a promising framework that unifies perception, reasoning, and control for robot manipulation by adapting pretrained vision-language models (VLMs) to action prediction. However, VLM-derived representations are often insensitive to subtle visual distinctions required for low-level control, causing state aliasing between visually similar states that require substantially different actions. Prior VLA studies improve visual understanding by generating visual or reasoning outputs, such as future frames, 2D grounding points or traces, or intermediate spatial reasoning steps, but these objectives typically shape the vision encoder only indirectly through end-to-end prediction and do not explicitly analyze state aliasing in the learned visual feature space. To mitigate state aliasing, we introduce inverse dynamics learning as an auxiliary objective that directly supervises the VLA vision encoder. By predicting the action between current and future observations, our objective encourages the encoder to capture fine-grained visual distinctions that determine low-level actions. We further use pseudo-reversed supervision to expose the encoder to a broader range of action directions and improve generalization under limited robot demonstrations. Our method applies to diverse VLA baselines, uses only standard observation-action pairs without additional annotations, and preserves the original inference pipeline at test time. Experiments on CALVIN ABC-D and SimplerEnv show consistent gains across diverse VLA baselines. Frozen-encoder probing and state-feature alignment analyses further show that our method learns state-discriminative visual representations that reduce state aliasing and better align with robot state changes.

Vision-Language Models (VLMs) face severe memory and latency bottlenecks due to high-resolution visual tokens. While current token reduction methods theoretically save FLOPs, post-hoc pruning introduces structural overhead, failing to yield proportional wall-clock acceleration. However, enforcing a contiguous compact pathway risks geometric disorientation and loss of fine-grained localization. To overcome these barriers, this paper introduces CIVIC, a path-consistent compact visual inference framework. By maintaining compact sequence representations seamlessly across the vision encoder, projection layer, LLM prefill, and KV-cache, CIVIC avoids non-contiguous memory access and localized unmerging overheads. Evaluated on the Qwen3-VL architecture, CIVIC successfully translates sequence reductions into genuine physical hardware efficiency, shrinking KV-cache memory to approximately one-third of the baseline and reducing end-to-end inference latency. Enabled by text-aligned KL distillation and an adaptive spatial retention floor, CIVIC achieves these efficiency milestones without degrading accuracy across rigorous multimodal reasoning and visual grounding benchmarks.

Existing video captioning methods struggle to balance visual fidelity and redundancy: holistic captions are compact but lose fine-grained evidence, whereas segment-wise captions improve coverage but introduce heavy redundancy. We propose CodecCap, a codec-inspired framework for high-fidelity dense video captioning. Analogous to video codecs, CodecCap represents videos using keyframe and residual captions. Keyframe captions exhaustively encode stable visual context, while residual captions capture temporally only localized actions, motions and changes. This effectively preserves fine-grained visual evidence while reducing redundant descriptions. To quantify the fidelity of captions, we introduce VidCapQA, a caption-then-QA benchmark with 1,000 questions across 14 capability dimensions. Results on VidCapQA show that captions directly generated by strong VLMs still miss many visual details, highlighting caption representation as a critical bottleneck. Experiments show that CodecCap significantly surpasses direct captioning with the same underlying VLMs, suggesting keyframe-residual captioning a way for high-fidelity video-language supervision. We further use CodecCap to construct CodecVDC-100K, a large-scale dense captioning dataset with anchor, residual, scene-level, and video-level supervision.

We introduce LLaVA-OneVision-2 (LLaVA-OV-2), the most capable vision-language model in the LLaVA-OneVision series to date, achieving superior performance across a broad range of multimodal benchmarks. The model builds on a native OneVision-Encoder and incorporates Windowed Attention for efficient local computation while maintaining native resolution. Its key advance is codec-stream tokenization: it treats compressed video as a continuous bit-cost stream, where bit-cost dynamics determine adaptive temporal groups, and motion-residual cues select salient spatial evidence into compact visual canvases. This allocation concentrates a limited token budget on event-bearing content, enabling more stable long-video token compression than fixed groups of pictures. A shared 3D RoPE further places codec canvases, sampled frames, and images in a unified spatiotemporal coordinate system. Furthermore, we build the LLaVA-OV-2 data and training stack around large-scale open supervision: approximately 8M re-captioned video samples for pretraining, a 4M-sample spatial corpus for fine-tuning. We also introduce JumpScore, a temporal-localization benchmark targeting fine-grained grounding in high-frequency, densely repeated motion, a regime underrepresented by existing video evaluations. A standout capability of LLaVA-OV-2 is its unified perception across video understanding, temporal grounding, spatial grounding, and manipulation-trace reasoning. On JumpScore, LLaVA-OneVision-2-8B reaches 74.9 JumpScore mAP, surpassing Qwen3-VL-8B (30.1) by +44.8 points; under matched visual-token budgets on the same benchmark, codec-stream inputs improve temporal grounding over frame sampling by +9.7 points. Across standard benchmarks, LLaVA-OneVision-2-8B further outperforms Qwen3-VL-8B by +4.3 average points on video tasks, +5.3 on spatial tasks, and +15.6 average J&F on tracking tasks.

Vision-Language Models such as CLIP exhibit strong zero-shot recognition capability by aligning images with textual concepts, yet they often underperform on multi-label recognition where multiple objects co-exist. A key bottleneck is that the [CLS] token, as a single global visual representation, is insufficient to faithfully encode diverse targets with varying scales, contexts, and co-occurrence patterns. To address this limitation, we present a new multi-label image recognition framework, termed PIAA, which formulates prediction as Patch-level Inference followed by Adaptive Aggregation. Specifically, we first enhance patch-wise predictions from two complementary perspectives: (i) mitigating semantic entanglement in the visual encoder to obtain more discriminative patch representations, and (ii) learning an unsupervised visual classifier to narrow the vision-language modality gap. We then introduce an adaptive aggregation module that consolidates patch-level scores into the final multi-label prediction. Notably, the entire pipeline is fully training-free, requiring no gradient updates or parameter fine-tuning. Experiments show that our method achieves strong improvements with minimal extra computation, exceeding a 6% mAP gain on the challenging NUS-WIDE benchmark over representative baselines. Code is available at https://github.com/akang-wang/PIAA.

Visual decoding from brain signals is a key challenge at the intersection of computer vision and neuroscience, requiring methods that bridge neural representations and computational models of vision. We introduce a tri-modal contrastive framework for EEG-based visual decoding that aligns EEG, visual, and textual representations within a unified latent space. Our approach follows a two-stage design. First, we pre-train an EEG encoder via masked reconstruction on unlabeled trials, learning spatio-temporal regularities that transfer robustly to downstream tasks. Second, we jointly align EEG, image, and LLM-generated textual descriptions through contrastive learning, where text supervision acts as a semantic regularizer that injects linguistic structure into the shared space without overwhelming the primary EEG-image signal. The encoder integrates subject-specific adaptation, graph-attention over channels, and temporal-spatial convolutional embeddings. On the Things-EEG2 200-way zero-shot benchmark, our framework achieves 54.1% Top-1 and 83.4% Top-5 accuracy, substantially exceeding the strongest prior baseline (32.4% / 64.0%), with paired Wilcoxon tests confirming significance (p < 0.01) over all in-subject baselines. We validate generalization on Things-MEG. Analysis reveals that compact embedding geometries (CN-CLIP) outperform much larger backbones, and that decoding aligns with established neurophysiology of visual processing. This work is a critical step towards robust, semantically-grounded visual decoding from non-invasive temporal neural signals. The source code is publicly available in https://github.com/anon-eeg/eeg_image_decoding.

Short-form video poses new challenges to the quality assessment of user-generated content (UGC) due to its complex generation pipeline, rapid content variation, and mixed distortions. To address this challenge, we propose an end-to-end video quality assessment (VQA) framework that employs a dense visual encoder based on CLIP, and incorporates compression priors derived from the frequency domain to generate artifact- and structure-aware weight maps for feature aggregation. By explicitly decomposing artifact, structure, and original visual feature branches and adaptively fusing them over time through a learned gating module, the proposed method achieves accurate and efficient quality prediction. Experimental results show that our method achieves strong performance on short-form video datasets in terms of average rank and linear correlation (SRCC: 0.736, PLCC: 0.787), while maintaining efficient inference runtime. The code and additional results are available at: https://github.com/xinyiW915/FGSVQA.

Occluded person re-identification focuses on matching partially visible pedestrians across multiple camera views. However, occlusions disrupt body-region cues, thereby complicating cross-view matching. Most person ReID methods built on pretrained vision-language models only focus on enhancing prompt-based feature learning while ignoring the semantic information of occluders. Based on the success of CLIP-ReID, we propose a novel Dual Prompt Learning ReID (DPL-ReID) model for occluded person ReID. It incorporates a Dual Prompt Learning (Dual-PL) strategy, which can utilize textual cues to capture complete pedestrian semantics and keep robustness against occlusion, and a Real-World Occlusion Augmentation (RWOA) method that realistically simulates occlusion scenarios encountered in real word to enrich occluded samples. In addition, we also design a Weighted Gated Feature Fusion (WGFF) method, which in corporates LSNet to capture global information and act as a feature-gating mechanism. This mechanism can effectively guide the CLIP visual encoder toward generating more comprehensive feature representations. Extensive experiments on several benchmark occluded ReID datasets show that our proposed DPL-ReID achieves the state-of-the art performance. The occlusion instance library are available at https://github.com/stone-qiao/DPL-ReID.

Modern multimodal large language models (MLLMs) typically keep the language model fixed and train a visual projector that maps the pixels into a sequence of tokens in its embedding space, so that images can be presented in essentially the same form as text. However, the language model has been optimized to operate on discrete, semantically meaningful tokens, while prevailing visual projectors transform an image into a long stream of continuous and highly correlated embeddings. This causes the visual tokens to behave differently from the word-like units that LLMs are originally trained to understand. We propose a novel Disentangled Visual Tokenization (DiVT) that clusters patch embeddings into coherent semantic units, so each token corresponds to a distinct visual concept instead of a rigid grid cell. DiVT further adapts its token budget to image complexity, providing an explicit accuracy-compute trade-off modifying neither the vision encoder nor the language model. Across diverse multimodal benchmarks, DiVT matches or surpasses baselines with significantly fewer visual tokens, demonstrating robustness under limited token budgets, significantly reducing memory cost and latency while making visual inputs more compatible with LLMs. Our code is available at https://github.com/snuviplab/DiVT.

The fundamental challenge in scaling Video Large Language Models (Video LLMs) to long-form video lies in managing the explosion of visual-token context length. Existing strategies predominantly focus on "post-hoc" token reduction -- reducing visual tokens after feature extraction to alleviate the LLM's computational overhead. While these methods effectively reduce the number of visual tokens, we observe that the primary latency bottleneck then shifts from the LLM to the expensive per-frame processing of the vision encoder. To address this, we introduce LiteFrame, a strong, yet highly efficient video encoder backbone for Video LLMs. To train LiteFrame, we propose Compressed Token Distillation (CTD), a novel training framework that teaches a compact student vision encoder to directly predict information-dense, spatio-temporally compressed representations produced by a large teacher vision model, effectively bypassing redundant computation. When coupled with further Language Model Adaptation (LMA), this approach results in a new latency-accuracy Pareto frontier -- compared with InternVL3-8B, LiteFrame provides a 35% reduction in end-to-end latency while processing 8$\times$ more frames and improves average video understanding accuracy across multiple benchmarks. Our results demonstrate a new potential path to unlocking longer-form video understanding under fixed compute budgets.

A central goal in understanding human vision is to uncover the visual features that drive neuronal activity. A growing body of work has used artificial neural networks as encoding models to predict cortical responses to natural images, revealing the visual content that activates category-selective regions. However, existing approaches are largely correlational and treat the encoder as a black box, leaving open which image features drive each voxel's response. We introduce Mechanistically Interpretable Neural Encoding (MINE), a framework that opens this black box by applying mechanistic-interpretability tools to localize the features within natural images that drive millimeter-scale (voxel-level) activity. MINE predicts each voxel's response using language-aligned image representations, and produces semantically interpretable descriptions of the features critical for the voxel's activation. We further generalize these per-image features into per-voxel functional profiles. To validate the per-image descriptions, we show they are sufficient to generate images that elicit voxel responses matching the responses to the original images, more accurately than images generated from random or low-attribution controls. Moreover, counterfactually inserting or removing the predicted features from images shifts activation in the expected direction, providing causal evidence. Counterfactual editing guided by the per-voxel activation profiles produces even stronger activation shifts, indicating that the profiles faithfully capture each voxel's selectivity. Finally, we apply MINE to well-studied category-selective brain regions, showing it recovers their known categorical preferences while revealing fine-grained unique voxel structure within each region. Overall, our results establish mechanistic interpretability as a path to discover and causally validate fine-grained hypotheses about neural function.

Reference-guided video editing takes a source video, a text instruction, and a reference image as inputs, requiring the model to faithfully apply the instructed edits while preserving original motion and unedited content. Existing methods fall into two paradigms, each with inherent limitations: decoupled encoders suffer from modality gaps when processing instructions and visual content independently, while unified vision-language encoders lose fine-grained spatial details by relying solely on final-layer representations. We observe that VLM layers encode complementary information hierarchically -- early layers capture localized spatial details essential for precise editing, while deeper layers encode global semantics for instruction comprehension. Building on this insight, we present MiVE (Multiscale Vision-language features for reference-guided video Editing), a framework that repurposes VLMs as multiscale feature extractors. MiVE extracts hierarchical features from Qwen3-VL and integrates them into a unified self-attention Diffusion Transformer, eliminating the modality mismatch inherent in cross-attention designs. Experiments demonstrate that MiVE achieves state-of-the-art performance by ranking highest in human preference, outperforming both academic methods and commercial systems.

Since the emergence of Vision Transformer (ViT), it has been widely used in generative language model and generative visual model. Especially in the current state-of-art open source multimodal models, ViT obtained by CLIP or SigLIP method serves as the vision encoder backbone to help them acquire visual understanding capabilities. But this method leads to limitations in visual understanding for details, such as difficulty in recognizing small text or numbers in images. To address these issues, we propose a new model to unify pix token and word token into the generative language model. The new model also features with each pix of image having its own token embedding, color folding, global conditional attention approximation and image unsupervised pretraining. We conducted image unsupervised pretraining experiments using our new model to explore its potential. The experimental results show that it has good performance even in small model and with limited training data. We believe our model also conforms to the scaling law, as long as model parameters and training data increased, its performance will continue to improve.

Brain encoder models predict cortical fMRI responses from the internal activations of pretrained vision and language networks, and are typically evaluated by held-out prediction accuracy. This is a useful signal for training but a poor one for interpretation: it tells us an encoder fits the data without telling us whether it has internalized the functional organization of the brain. We propose feature visualization -- gradient ascent on the encoder's predicted activation for a target region of interest (ROI) -- as a complementary interpretability technique, and apply it to TRIBE v2 composed with V-JEPA 2 (ViT-G, 40 layers), holding both frozen and synthesizing still images for seven regions spanning the ventral and dorsal visual hierarchies. Under identical hyperparameters, the probe recovers a visible progression of increasing spatial scale and feature complexity across V1 to V4, matching the ventral-stream hierarchy. It also produces three distinctive downstream regimes: radial "frozen-motion" streaks for the middle temporal area (MT) despite static-only optimization, face-like features for the fusiform face area (FFA), and consistent rectilinear line patterns for the parahippocampal place area (PPA). Optimized FFA stimuli drive the predicted region ~4x as much as a natural face photograph, consistent with feature visualization producing adversarial super-stimuli rather than canonical exemplars. The probe is simple, differentiable, and applicable to any brain encoder with a differentiable backbone, allowing for qualitative evaluation of brain encoders.

Text-guided inpainting has made image forgery increasingly realistic, challenging both SID and IFL. However, existing methods often struggle to point out suspicious signals across domains. To address this problem, we propose EDGER, a patch-based, dual-branch framework that localizes manipulated regions in arbitrary resolution images without sacrificing native resolution. The first branch, Edge-Guided Segmentation, introduces a Frequency-based Edge Detector to emphasize high-frequency inconsistencies at manipulation boundaries, and fine-tunes a SegFormer to fuse RGB and edge features for pixel-level masks. Since edge evidence is most informative only when patches contain both authentic and manipulated pixels, we complement Edge-Guided Segmentation with a Synthetic Heatmapping branch, a classification-based localizer that fine-tunes a CLIP-ViT image encoder with LoRA to flag fully synthetic patches. Together, Synthetic Heatmapping provides coarse, patch-level synthetic priors, while Edge-Guided Segmentation sharpens boundaries within partially manipulated patches, yielding comprehensive localization. Evaluated in the MediaEval 2025, SynthIM challenge, Manipulated Region Localization Task's setting, our approach scales to multi-megapixel imagery and exhibits strong cross-domain generalization. Extensive ablations highlight the complementary roles of frequency-based edge cues and patch-level synthetic priors in driving accurate, resolution-agnostic localization.

Vision-Language-Action (VLA) models have shown remarkable promise in robotics manipulation, yet their high computational cost hinders real-time deployment. Existing token pruning methods suffer from a fundamental trade-off: aggressive compression using pruning inevitably discards critical geometric details like contact points, leading to severe performance degradation. This forces a compromise, limiting the achievable compression rate and thus the potential speedup. We argue that breaking this trade-off requires rethinking compression as a geometry-aware, continuous token resampling in the vision encoder. To this end, we propose the Differentiable Grid Sampler (GridS), a plug-and-play module that performs task-aware, continuous resampling of visual tokens in VLA. By adaptively predicting a minimal set of salient coordinates and extracting features via differentiable interpolation, GridS preserves essential spatial information while achieving drastic compression (with fewer than 10% original visual tokens). Experiments on both LIBERO benchmark and a real robotic platform demonstrate that validating the lowest feasible visual token count reported to date, GridS achieves a 76% reduction in FLOPs with no degradation in the success rate. The code is available at https://github.com/Fediory/Grid-Sampler.

Vision-language models (VLMs), such as CLIP and SigLIP 2, are widely used for image classification, yet their vision encoders remain vulnerable to systematic biases that undermine robustness. In particular, correlations between foreground objects and their backgrounds constitute a salient and practically important class of spurious dependencies. In this work, we revisit the well-known property of high linear additivity in VLM embedding spaces and show that it enables a decomposition of scene representations into foreground and background components. Leveraging this insight, we introduce a pre-training approach that exploits this property to construct background-invariant representations using synthetic data. Our method achieves, to our knowledge, the first worst-group accuracy exceeding $90\%$ on Waterbirds under perfect ($100\%$) spurious correlation (i.e., no minority-group examples in the training data). Furthermore, it demonstrates strong sim-to-real transfer and requires no access to real-world debiased data, making it practical for real-world deployment.

Representation autoencoders that reuse frozen pretrained vision encoders as visual tokenizers have achieved strong reconstruction and generation quality. However, existing methods universally extract features from only the last encoder layer, discarding the rich hierarchical information distributed across intermediate layers. We show that low-level visual details survive in the last layer merely as attenuated residuals after multiple layers of semantic abstraction, and that explicitly fusing multi-layer features can substantially recover this lost information. We propose DRoRAE (Depth-Routed Representation AutoEncoder), a lightweight fusion module that adaptively aggregates all encoder layers via energy-constrained routing and incremental correction, producing an enriched latent compatible with a frozen pretrained decoder. A three-phase decoupled training strategy first learns the fusion under the implicit distributional constraint of the frozen decoder, then fine-tunes the decoder to fully exploit the enriched representation. On ImageNet-256, DRoRAE reduces rFID from 0.57 to 0.29 and improves generation FID from 1.74 to 1.65 (with AutoGuidance), with gains also transferring to text-to-image synthesis. Furthermore, we uncover a log-linear scaling law ($R^2{=}0.86$) between fusion capacity and reconstruction quality, identifying \textit{representation richness} as a new, predictably scalable dimension for visual tokenizers analogous to vocabulary size in NLP.

Precise spatial reasoning is fundamental to robotic manipulation, yet the visual backbones of current vision-language-action (VLA) models are predominantly pretrained on 2D image data without explicit 3D geometric supervision, resulting in representations that lack accurate spatial awareness. Existing implicit spatial grounding methods partially address this by aligning VLA features with those of 3D-aware foundation models, but they rely on empirical layer search and perform alignment on LLM-level visual tokens where spatial structure has already been entangled with linguistic semantics, limiting both generalizability and geometric interpretability. We propose VEGA (Visual Encoder Grounding Alignment), a simple yet effective framework that directly aligns the output of the VLA's visual encoder with spatially-aware features from DINOv2-FiT3D, a DINOv2 model fine-tuned with multi-view consistent 3D Gaussian Splatting supervision. By performing alignment at the visual encoder output level, VEGA grounds spatial awareness before any linguistic entanglement occurs, offering a more interpretable and principled alignment target. The alignment is implemented via a lightweight projector trained with a cosine similarity loss alongside the standard action prediction objective, and is discarded at inference time, introducing no additional computational overhead. Extensive experiments on simulation benchmark and real-world manipulation tasks demonstrate that VEGA consistently outperforms existing implicit spatial grounding baselines, establishing a new state-of-the-art among implicit spatial grounding methods for VLA models.

The swift advancement in photo-realistic face generation technology has sparked considerable concerns across society and academia, emphasizing the requirement of generalizable face forgery detection and localization methods. Prior works tend to capture face forgery patterns across multiple domains using image modality, other modalities like fine-grained texts are not comprehensively investigated, which restricts the generalization capability of models. Besides, they usually analyze facial images created by GAN, but struggle to identify and localize those synthesized by diffusion. To solve the problems, in this paper, we devise a novel multi-domain fine-grained vision-language reconstruction (MFVLR) model, which explores comprehensive and diverse visual forgery traces via language-guided face forgery representation learning, to achieve generalizable diffusion-synthesized face forgery detection and localization (DFFDL). Specifically, we devise a fine-grained language transformer that studies general fine-grained language embeddings using language reconstruction. We propose a multi-domain vision encoder to capture general and complementary visual forgery patterns across the image and residual domains. A vision decoder is designed to reconstruct image appearance and achieve forgery localization. Besides, we propose an innovative plug-and-play vision injection module to enhance the interaction between the vision and language embeddings. Extensive experiments and visualizations demonstrate that our network outperforms the state of the art on different settings like cross-generator, cross-forgery, and cross-dataset evaluations.

Multilingual vision-language models exhibit systematic performance gaps across languages, but the mechanism remains ambiguous: cross-language divergence could arise from the visual encoder, the text branch, or their interaction. We resolve this ambiguity through a dense multilingual CLIP probe in which the visual encoder is held identical across thirteen typologically diverse languages and only the XLM-RoBERTa text branch varies. We evaluate two CLIP architectures spanning a 7x visual-encoder scale gap (XLM-R base + ViT-B/32, ~87M visual parameters; XLM-R large + ViT-H/14, ~632M) on 11 concepts and 210 images, and quantify cross-language agreement via cluster-mask IoU, top-percentile IoU, and Spearman rank correlation against an English reference (n=2,310 paired observations per language). Three findings emerge. First, low-resource languages (Arabic, Basque, Luxembourgish) incur a structural penalty at both backbone scales (Wilcoxon HR>LR p<10^-300; cluster-mask IoU gap +0.114 at base, +0.143 at large), isolating the deficit to the text branch. Second, scaling the encoder 7x widens the gap for structural failure cases (Basque Δ=-0.056, Luxembourgish Δ=-0.076) while improving Arabic (Δ=+0.033), separating corpus-coverage from tokeniser-fertility failures. Third, peak similarity is preserved across languages (mean ratio 0.94 at large scale) while cluster-mask IoU drops sharply, identifying spatial misalignment, not signal collapse, as the dominant failure mode. At 3.4-3.9 Wh per 1,000 queries, dense-CLIP grounding is competitive with high-throughput inference budgets, positioning it as a practical substrate for energy-aware multilingual deployment.

Visual encoding constitutes a major computational bottleneck in Multimodal Large Language Models (MLLMs), especially for high-resolution image inputs. The prevailing practice typically adopts global encoding followed by post-ViT compression. Global encoding produces massive token sequences, while post-ViT compression incurs the full quadratic attention cost of the ViT before any token reduction takes place. In this work, we revisit this convention along two dimensions: the encoding strategy and visual token compression. First, controlled experiments show that slice-based encoding outperforms global encoding across benchmarks, suggesting that preserving local details through sliced views can be more beneficial than applying global attention for fine-grained perception. Second, we introduce intra-ViT early compression, which reduces tokens in shallow ViT layers and substantially lowers visual-encoding FLOPs while preserving downstream performance. By integrating intra-ViT compression into the slice-based encoding framework, we present LLaVA-UHD v4, an efficient and compute-controllable visual encoding scheme tailored for high-resolution inputs. Across a diverse set of benchmarks covering document understanding, OCR, and general VQA, LLaVA-UHD v4 reduces visual-encoding FLOPs by 55.8% while matching or even surpassing baseline performance. These results suggest that visual-encoding efficiency can be substantially improved without sacrificing downstream performance, providing a practical design direction for efficient high-resolution MLLMs. All model weights and code will be publicly released to support further research.

Current image quality assessment methods are heavily biased towards global distortions (e.g., noise, blur), neglecting local perceptual artifacts such as ghosting, lens flare, and moire effects. Although significant progress has been made in artifact removal, the fundamental problem of automatic artifact detection remains largely unexplored. In this paper, we formalize the Image Perceptual Artifact Detection (IPAD) task to address this gap. We contribute a benchmark dataset comprising 3,520 artifact images, including 520 real-captured and 3,000 synthetic samples, each paired with pixel-level masks across three representative artifact categories. The core challenge of IPAD lies in the localized, subtle, and semantically weak nature of these artifacts, which makes them prone to missed detection. To overcome this, we introduce IPAD-CLIP, a novel framework built upon CLIP that enhances artifact discrimination in both textual and visual spaces while preserving generalization capabilities. Our key insight is that local artifacts often exhibit strong correlations with specific semantic contexts. Accordingly, we learn artifact-aware text embeddings to explicitly model the object-artifact relationships, resulting in enhanced representations that clear differentiate between clean and artifact prompts. These text embeddings are then used as anchors to shift the visual encoder's attention from high-level semantics to subtle, low-level artifacts. Extensive experiments demonstrate that IPAD-CLIP offers a resource-efficient adaptation of CLIP for detection, significantly outperforming advanced image anomaly detection and manipulation detection methods on our benchmark. To the best of our knowledge, this is the first study addressing multi-class local perceptual artifact detection in terms of both dataset and model.

We present ZAYA1-VL-8B, a compact mixture-of-experts vision-language model built upon our in-house language model, ZAYA1-8B. Despite its compact size, ZAYA1-VL achieves performance competitive with leading base models such as Molmo2-4B and InternVL3.5-4B, while surpassing models including Qwen2.5-VL-3B, PLM-3B, and MolmoE-1B across a range of image understanding, reasoning, and counting benchmarks. The architecture incorporates two key innovations: (1) vision-specific LoRA adapters integrated into the LLM to increase modality-specific capacity without increasing the number of experts, and (2) bidirectional attention over image tokens within the LLM to enhance visual understanding. We detail the full training pipeline including data composition at each stage, sequence packing, and the attention masking scheme. The model comprises 9.2B total parameters, with 1.4B active parameters including the vision encoder, and is publicly available at https://huggingface.co/Zyphra/ZAYA1-VL.

Vision-language models (VLMs) raise growing concerns about privacy, copyright, and bias, motivating machine unlearning to remove sensitive knowledge. However, existing methods primarily fine-tune the language decoder, leading to superficial forgetting that fails to erase underlying visual representations and often introduces object hallucination. We propose HFRU, a reinforcement unlearning framework that operates on the vision encoder for deep semantic removal. Our two-stage approach combines alignment disruption with GRPO-based optimization using a composite reward, including an abstraction reward that encourages semantically valid substitutions and mitigates hallucinations. Experiments on object recognition and face identity tasks show that HFRU achieves over 98% forgetting and retention performance, while introducing negligible object hallucination, significantly outperforming prior methods.Our code and implementation details are available at https://github.com/XMUDeepLIT/HFRU.

Semantic segmentation of large-scale 3D point clouds is crucial for applications such as autonomous driving and urban digital twins. However, the sparse sampling pattern of LiDAR and the view-dependent geometric distortion in image observations complicate cross-modal alignment and hinder stable fusion. Inspired by the fact that 2D images captured by cameras are representations of the 3D world, we recognize that the features learned from 2D and 3D segmentation share some common semantics, while other aspects remain modality-specific. This insight motivates a unified multimodal framework for joint 2D-3D semantic segmentation. We combine a SAM-based vision encoder with a SPTNet-based geometric encoder to extract complementary semantic and geometric representations. The resulting features from both modalities are explicitly decomposed into shared and private subspaces, where the shared components summarize semantic factors common to both domains, and the private components preserve properties that are unique to each modality. A lightweight attention-based fusion module aggregates the shared features into a consistent cross-modal representation, and a regularized training objective ensures both semantic alignment and subspace independence. Experiments on the SemanticKITTI and nuScenes benchmarks demonstrate consistent improvements in segmentation accuracy over representative multimodal baselines, accompanied by competitive computational efficiency. Cross-domain evaluation on nuScenes USA-Singapore shows stable performance under distribution shifts, demonstrating strong generalization. The implementation code is publicly available at: https://github.com/shuaizhang69/UniD-Shift.

Vision Language Models (VLMs) offer the exciting possibility of processing text as rendered images, bypassing the need for tokenizing the text into long token sequences. Since VLM image encoders map fixed-size images to a fixed number of visual tokens, varying rendering resolution provides a fine-grained compression knob. However, accuracy deteriorates quickly as compression increases: characters shrink below the vision encoder's effective resolution, making them indistinguishable. To address this, we propose LensVLM, an inference framework and post-training recipe that enables VLMs to scan compressed images, then selectively expand only the relevant images to their uncompressed form via learned tools. Building on Qwen3.5-9B-Base, LensVLM maintains accuracy comparable to the full-text upper bound at 4.3x effective compression and outperforms retrieval-based, text- and visual-compression baselines up to 10.1x effective compression across seven text QA benchmarks. LensVLM also generalizes to multimodal document and code understanding tasks, with the accuracy gain over baselines growing as compression increases. Our analysis validates this approach: training makes visual compression robust to rendering choices, and as compression grows the model increasingly relies on expanded content rather than unreliable visual reading. The analysis also yields practical tool-choice guidance: text expansion is preferable for rendered text, while high-resolution image expansion suits native documents whose layout cues carry task-relevant information.

Video semantic search in densely crowded scenes remains a challenging task due to visual encoders tendency to prioritize salient foreground regions while neglecting contextually important, background areas. We propose an Inverse Attention Embedding mechanism that explicitly captures and highlights these overlooked regions. By combining inverse attention embeddings with traditional visual embeddings, our method significantly enhances semantic retrieval performance without additional training. Initial experiments and ablation studies demonstrate promising improvements over existing approaches in recall for video semantic search in crowded environments.

Zero-shot 3D anomaly detection aims to identify anomalies without access to training data from target categories. However, existing methods mainly rely on projecting 3D observations into multi-view representations that primarily capture geometric cues rather than realistic visual semantics and process them with vision encoders pretrained on RGB data, leading to a significant domain gap between the encoder and the projected representations. To address this issue, we propose Align3D-AD, a unified two-stage framework that leverages the RGB modality from auxiliary categories as cross-modal guidance for zero-shot 3D anomaly detection. First, we introduce a cross-modal feature alignment paradigm that maps rendering features into the RGB semantic space. Unlike prior works that implicitly rely on pretrained encoders, our method enables direct semantic transfer from RGB observations. A semantic consistency reweighting strategy is further introduced to refine feature alignment by reweighting local regions according to holistic semantic consistency. Second, we propose a modality-aware prompt learning framework with dual-prompt contrastive alignment. By assigning independent prompts to RGB-aligned and rendering features, our method captures complementary semantics across modalities, while the contrastive alignment further enhances prompt representations to improve discriminability. Extensive experiments on MVTec3D-AD, Eyecandies, and Real3D-AD demonstrate that Align3D-AD consistently outperforms existing zero-shot methods under both one-vs-rest and cross-dataset settings, highlighting its generalization capability and robustness. Code and the dataset will be made available once our paper is accepted.

Video editing has evolved toward In-Context Learning (ICL) paradigms, yet the resulting quadratic attention costs create a critical computational bottleneck. In this work, we propose In-context Sparse Attention (ISA), the first near-lossless empirical sparse framework tailored for ICL video editing. Our design is grounded in two key insights: first, context tokens exhibit significantly lower saliency than source tokens; second, we theoretically prove and empirically validate that Query sharpness correlates with approximation error. Motivated by these findings, ISA implements an efficient pre-selection strategy to prune redundant context, followed by a dynamic query grouping mechanism that routes high-error queries to full attention and low-error ones to a computationally efficient 0-th order Taylor sparse attention. Furthermore, we build \textbf{\texttt{LIVEditor-14B}} , a novel lightning video editing model via ISA and a proposed video-editing data pipeline that curated a 1.7M high-quality dataset. Extensive experiments demonstrate that LIVEditor-14B achieves a $\sim$60% reduction in attention-module latency while surpassing state-of-the-art methods across EditVerseBench, IVE-Bench, and VIE-Bench, delivering near-lossless acceleration without compromising visual fidelity.

Recent advances in omni-modal large language models have enabled remarkable progress in joint vision-audio understanding. However, prevailing architectures rely on modality-specific encoders with a \emph{video-coarse, audio-dense} design -- sampling visual frames at 1--2 fps while processing audio waveforms at 25 fps -- resulting in systems that perceive video \emph{frame by frame, modality by modality} rather than holistically as humans do. Such a discrepancy leaves models with impoverished cross-modal interaction during encoding and an inability to capture fine-grained visual motion. To bridge this gap, we present \textbf{Omni-Encoder, a unified Transformer backbone designed to co-embed visual and audio signals at a symmetrical 25 fps} within a shared latent space. This architecture leverages three core innovations -- the Omni-Encoder Token Template, Omni-RoPE, and Temporal Window Shifting -- to effectively reconcile the dual challenges of modality disentanglement and computational efficiency. Experiments demonstrate that, compared to the modality-specific baseline Qwen2.5-Omni under the same input token budget to the LLM decoder, Omni-Encoder delivers substantial gains on visual continuous understanding tasks -- such as sign language recognition and fine-grained sports action analysis -- while maintaining competitive performance on established audio-visual benchmarks such as AVQA and Speaker Identification and Localization. These results suggest that unified omnivorous encoding offers a promising direction for building omni-modal models that more closely reflect the integrated nature of human perception.

Vision-Language Models (VLMs) have enhanced traditional LLMs with visual capabilities through the integration of vision encoders. While recent works have explored various combinations of vision encoders and LLMs, there still lacks a principled understanding of what makes a vision encoder suitable for VLM alignment. In this paper, we systematically investigate this question via comprehensive experiments on a curated collection of 19 pre-trained vision encoders from diverse sources. We first demonstrate that common practices, such as choosing encoders with the largest size or highest zero-shot accuracy, consistently fail to identify optimal models. In fact, these metrics show only weak to moderate correlation with VLM performance. This intriguing finding begs a fundamental question: What factors of vision-encoders matter in VLM? Through comprehensive analysis, we identify that the structural similarity across modalities plays a crucial but previously overlooked role in vision-encoder selection, which we measure using the Gromov-Wasserstein distance as a proxy. From a theoretical perspective, we show that the learnability of cross-modality mapping can be provably associated with the Gromov-Wasserstein distance. Empirical verification on 60+ full VLM training runs shows that our proposed inference-only metric performs significantly better than alternative model selection strategies and exhibits a much stronger correlation with final VLM performance, thereby enabling efficient and effective prediction of VLM performance before full training.

In this paper, we present \textbf{Gen}erative \textbf{L}anguage-\textbf{I}mage \textbf{P}re-training (GenLIP), a minimalist generative pretraining framework for Vision Transformers (ViTs) designed for multimodal large language models (MLLMs). To better align vision encoders with the autoregressive nature of LLMs, GenLIP trains a ViT to predict language tokens directly from visual tokens using a standard language modeling objective, without contrastive batch construction or an additional text decoder. This design offers three key advantages: (1) \textbf{Simplicity}: a single transformer jointly models visual and textual tokens; (2) \textbf{Scalability}: it scales effectively with both data and model size; and (3) \textbf{Performance}: it achieves competitive or superior results across diverse multimodal benchmarks. Trained on 8B samples from Recap-DataComp-1B, GenLIP matches or surpasses strong baselines despite using substantially less pretraining data. After continued pretraining on multi-resolution images at native aspect ratios, GenLIP further improves on detail-sensitive tasks such as OCR and chart understanding, making it a strong foundation for vision encoders in MLLMs.

In this paper, we propose an HOI-aware adaptive network named AdaAct for weakly-supervised action segmentation. Most existing methods learn a fixed network to predict the action of each frame with the neighboring frames. However, this would result in ambiguity when estimating similar actions, such as pouring juice and pouring coffee. To address this, we aim to exploit temporally global but spatially local human-object interactions (HOI) as video-level prior knowledge for action segmentation. The long-term HOI sequence provides crucial contextual information to distinguish ambiguous actions, where our network dynamically adapts to the given HOI sequence at test time. More specifically, we first design a video HOI encoder that extracts, selects, and integrates the most representative HOI throughout the video. Then, we propose a two-branch HyperNetwork to learn an adaptive temporal encoder, which automatically adjusts the parameters based on the HOI information of various videos on the fly. Extensive experiments on two widely-used datasets including Breakfast and 50Salads demonstrate the effectiveness of our method under different evaluation metrics.

Brain encoding models not only serve to decipher how visual stimuli are transformed into neural responses, but also represent a critical step toward visual prostheses that restore vision for patients with severe vision disorders. Brain encoding involves two fundamental steps: achieving faithful reconstruction of neural responses and establishing cross-modal alignment between visual stimuli and neural responses. To this end, we propose ViBE, a novel brain encoding framework for generating magnetoencephalography (MEG) and electroencephalography (EEG) signals from visual stimuli. Specifically, we first design a spatio-temporal convolutional variational autoencoder (TSC-VAE) that captures the spatio-temporal characteristics of M/EEG signals for effective neural response reconstruction. To bridge the modality gap between visual features and neural representations, we employ Q-Former to map CLIP image embeddings to the TSC-VAE latent space, producing neural proxy embeddings. For comprehensive cross-modal alignment, we combine mean squared error (MSE) loss for point-wise feature matching with sliced Wasserstein distance (SWD) for probability distribution alignment between the neural proxy embeddings and TSC-VAE latent embeddings. We conduct extensive experiments on the THINGS-EEG2 and THINGS-MEG datasets, demonstrating the effectiveness of our approach in generating high-quality M/EEG signals from visual stimuli.

We present LatentDiff, a scalable framework for semantic dataset comparison that operates directly in the latent space of pretrained vision encoders. By combining sparse autoencoder-based divergence testing with density ratio estimation, LatentDiff identifies interpretable semantic differences between datasets at a fraction of the computational cost of caption-based alternatives. We also introduce Noisy-Diff, a benchmark capturing realistic sparse distribution shifts that cause existing methods to struggle. Experiments demonstrate that LatentDiff achieves superior accuracy while remaining robust to settings where an extremely small fraction of images (from 5% to <1% ) differ semantically.

Vision foundation models (VFMs), such as DINOv3, provide rich semantic representations that are promising for computational pathology. However, many current adaptations pair frozen VFMs with lightweight decoders, creating a capacity mismatch that often limits boundary fidelity for infiltrative tumor bulk segmentation. This paper presents Dino-NestedUNet, a framework that couples a pre-trained DINOv3 encoder with a Nested Dense Decoder. Instead of sparse skip connections and linear upsampling, the proposed decoder forms a dense grid of intermediate pathways to enable continuous feature reuse and multi-scale recalibration, aligning high-level semantics with low-level morphological textures during reconstruction. We evaluate Dino-NestedUNet on three histopathology cohorts (multi-center CHTN, institutional OSU, and CAMELYON16) and observe consistent improvements over UNet++ and standard Dino-UNet variants, particularly under cross-domain shift. To further assess external generalization, we perform zero-shot evaluation by training on CHTN and directly testing on unseen TIGER WSIBULK and OSU CRC cohorts without fine-tuning. These results suggest that dense decoding is a key ingredient for unlocking foundation encoders in boundary-sensitive pathology segmentation.

Unified multimodal models typically rely on pretrained vision encoders and use separate visual representations for understanding and generation, creating misalignment between the two tasks and preventing fully end-to-end optimization from raw pixels. We introduce Tuna-2, a native unified multimodal model that performs visual understanding and generation directly based on pixel embeddings. Tuna-2 drastically simplifies the model architecture by employing simple patch embedding layers to encode visual input, completely discarding the modular vision encoder designs such as the VAE or the representation encoder. Experiments show that Tuna-2 achieves state-of-the-art performance in multimodal benchmarks, demonstrating that unified pixel-space modelling can fully compete with latent-space approaches for high-quality image generation. Moreover, while the encoder-based variant converges faster in early pretraining, Tuna-2's encoder-free design achieves stronger multimodal understanding at scale, particularly on tasks requiring fine-grained visual perception. These results show that pretrained vision encoders are not necessary for multimodal modelling, and end-to-end pixel-space learning offers a scalable path toward stronger visual representations for both generation and perception.

Vision Transformers (ViTs) dominate self-supervised learning (SSL). While they have proven highly effective for large-scale pretraining, they are computationally inefficient and scale poorly with image size. Consequently, foundational models like DINO are constrained to low-resolution processing. A recent foveal-inspired transformer achieves resolution agnosticism by iteratively processing a fixed-size context of multi-zoom patches. This model demonstrated promising results via supervised learning, utilizing a sequential, recurrent-like process without backpropagation through time. To unlock its potential as a foundational backbone, we introduce a novel sequential-to-global SSL framework based on DINO's self-distillation objective. Supported by an efficient integral-image patch extraction method, our approach enables large-scale pretraining for image-size agnostic vision encoders. We achieve competitive performance on ImageNet-1K and downstream classification tasks, maintaining a constant computational budget regardless of input resolution.

Early action prediction seeks to anticipate an action before it fully unfolds, but limited visual evidence makes this task especially challenging. We introduce EAST, a simple and efficient framework that enables a model to reason about incomplete observations. In our empirical study, we identify key components when training early action prediction models. Our key contribution is a randomized training strategy that samples a time step separating observed and unobserved video frames, enabling a single model to generalize seamlessly across all test-time observation ratios. We further show that joint learning on both observed and future (oracle) representations significantly boosts performance, even allowing an encoder-only model to excel. To improve scalability, we propose a token masking procedure that cuts memory usage in half and accelerates training by 2x with negligible accuracy loss. Combined with a forecasting decoder, EAST sets a new state of the art on NTU60, SSv2, and UCF101, surpassing previous best work by 10.1, 7.7, and 3.9 percentage points, respectively.

Continuous inference over concurrent video streams imposes substantial compute and memory demands on vision-language model (VLM) serving. Streaming inference uses sliding windows to maintain a bounded context of recent video, but processing each window independently repeats visual encoding and large language model (LLM) prefilling for similar and overlapping content. Existing optimizations provide limited coordination across these stages and often rely on model-specific training, profiling, or model-generated signals. We present CodecSight, a streaming VLM serving system that uses codec metadata as shared runtime guidance across visual encoding and LLM prefilling, without model-specific training or offline profiling. Codec-derived change signals guide patch pruning before visual encoding, reducing both visual computation and the number of downstream visual tokens. Codec-defined frame types guide selective key-value (KV) refresh across windows, while positional correction enables reuse of the remaining cached keys. Across three VLMs and four video workloads, our vLLM-based implementation supports up to $3.3\times$ as many concurrent streams and achieves up to a $5.3\times$ speedup in average time-to-first-token relative to the state-of-the-art baselines. It also reduces executed FLOPs by up to 93%, with a maximum task-quality decrease of 4.64 percentage points.

Pretrained Vision Transformers (ViTs) such as DINOv2 and MAE provide generic image features that can be applied to a variety of downstream tasks such as retrieval, classification, and segmentation. However, such representations tend to focus on the most salient visual cues in the image, with no way to direct them toward less prominent concepts of interest. In contrast, Multimodal LLMs can be guided with textual prompts, but the resulting representations tend to be language-centric and lose their effectiveness for generic visual tasks. To address this, we introduce Steerable Visual Representations, a new class of visual representations, whose global and local features can be steered with natural language. While most vision-language models (e.g., CLIP) fuse text with visual features after encoding (late fusion), we inject text directly into the layers of the visual encoder (early fusion) via lightweight cross-attention. We introduce benchmarks for measuring representational steerability, and demonstrate that our steerable visual features can focus on any desired objects in an image while preserving the underlying representation quality. Our method also matches or outperforms dedicated approaches on anomaly detection and personalized object discrimination, exhibiting zero-shot generalization to out-of-distribution tasks.

Computational pathology needs whole-slide image (WSI) foundation models that transfer across diverse clinical tasks, yet current approaches remain largely slide-centric, often depend on private data and expensive paired-report supervision, and do not explicitly model relationships among multiple slides from the same patient. We present MOOZY, a patient-first pathology foundation model in which the patient case, not the individual slide, is the core unit of representation. MOOZY explicitly models dependencies across all slides from the same patient via a case transformer during pretraining, combining multi-stage self-supervision with scaled low-cost task supervision. In Stage 1, we pretrain a vision-only slide encoder on 77,134 public slide feature grids using masked self-distillation. In Stage 2, we align these representations with clinical semantics using a case transformer and multi-task supervision over 333 tasks from 56 public datasets, including 205 classification and 128 survival tasks across four endpoints. Across sixteen held-out tasks, MOOZY improves macro weighted F1, balanced accuracy, and macro weighted ROC-AUC relative to PRISM by +4.19%, +7.93%, and +6.95%, respectively. MOOZY is also parameter efficient with 85.77M parameters, 14$\times$ smaller than GigaPath. These results suggest that patient-level pretraining yields transferable embeddings, providing a path toward scalable patient-first histopathology foundation models.

We present UniMotion, to our knowledge the first unified framework for simultaneous understanding and generation of human motion, natural language, and RGB images within a single architecture. Existing unified models handle only restricted modality subsets (e.g., Motion-Text or static Pose-Image) and predominantly rely on discrete tokenization, which introduces quantization errors and disrupts temporal continuity. UniMotion overcomes both limitations through a core principle: treating motion as a first-class continuous modality on equal footing with RGB. A novel Cross-Modal Aligned Motion VAE (CMA-VAE) and symmetric dual-path embedders construct parallel continuous pathways for Motion and RGB within a shared LLM backbone. To inject visual-semantic priors into motion representations without requiring images at inference, we propose Dual-Posterior KL Alignment (DPA), which distills a vision-fused encoder's richer posterior into the motion-only encoder. To address the cold-start problem -- where text supervision alone is too sparse to calibrate the newly introduced motion pathway -- we further propose Latent Reconstruction Alignment (LRA), a self-supervised pre-training strategy that uses dense motion latents as unambiguous conditions to co-calibrate the embedder, backbone, and flow head, establishing a stable motion-aware foundation for all downstream tasks. UniMotion achieves state-of-the-art performance across seven tasks spanning any-to-any understanding, generation, and editing among the three modalities, with especially strong advantages on cross-modal compositional tasks.