Junfeng Wang, Jiawei Liu, Yongchao Xu, Tao Jiang, Jiangbo Ai, Jin Zhang, Chong Wang, Qixing Zhang
Complex visual question answering often requires vision-language models to think with images, i.e., actively acquire local visual evidence by zooming into informative regions of high-resolution images. Our error analysis shows that many failures in this setting originate from incorrect observation actions, where the model fails to localize key regions or obtain crops that contain sufficient evidence for answering. We further observe that bounding-box tokens exhibit substantially higher generation entropy than ordinary language tokens, indicating high uncertainty when the model decides where to look. However, existing reinforcement learning methods based on final-answer rewards provide only trajectory-level feedback, making it difficult to assign precise credit to the observation actions. We propose Observation Policy Optimization (OPO), an RL algorithm for optimizing observation actions in tool-augmented thinking with images. OPO consists of two core modules. First, Uncertainty-Guided Observation Branching (UOB) treats each image zoom-in tool call as an observation event and selectively branches at high-uncertainty, low-redundancy events during rollout, enabling efficient exploration of alternative cropping decisions and their resulting visual evidence. Second, Evidence-Localized Advantage Attribution (ELAA) compares sibling branches using local evidence-sufficiency rewards and assigns credit only to the corresponding observation-action tokens. By decoupling observation-action optimization from trajectory-level outcome supervision, OPO improves visual evidence acquisition and further enhances visual reasoning performance. Experiments show that OPO consistently improves Qwen3-VL models across scales and achieves competitive performance against larger models and tool-augmented baselines. Code is publicly available.
We present FLEG, a feed-forward network that reconstructs language-embedded 3D Gaussians from multi-view images. Previous feed-forward language-embedded Gaussian reconstruction methods are restricted to a fixed number of input views and typically attach a language-aligned semantic embedding to each Gaussian, resulting in impractical input settings and semantic redundancy. In contrast, we introduce a geometry-semantic dual-branch distillation framework that supports a variable number of uncalibrated and unposed multi-view images as input. We also propose a novel-view-based distillation strategy during training that mitigates overfitting to input views. In addition, we observe that semantic representations are significantly sparser than geometric ones, and per-Gaussian language embedding is unnecessary. To exploit this sparsity, we design a decoupled language embedding strategy that represents language information with a sparse set of semantic Gaussians, rather than attaching embeddings to every Gaussian. Compared with dense pixel-aligned per-Gaussian embedding schemes, our method uses only 5% of the language embeddings while maintaining comparable semantic fidelity, effectively reducing storage costs. Extensive experiments demonstrate that FLEG outperforms state-of-the-art feed-forward reconstruction and language-embedded Gaussian methods in both reconstruction quality and language-aligned semantic representation.
Yunpeng Gong, Can Yang, Qingyuan Zeng, Dejun Xu, Zhiming Luo, Zhenzhong Wang, Min Jiang
Cross-modal person re-identification (ReID) adversarial attacks face a transferability challenge: existing methods may exhibit source-pair overfitting, where learned perturbations become overly dependent on the feature statistics of the source modality pair, thereby limiting their transferability under heterogeneous domain shifts. We argue that a key challenge is reducing source-pair-specific bias while maintaining attack effectiveness. To address this, we propose Multiform Attack (MA). The framework first learns a universal perturbation via Mahalanobis-guided gradient optimization. Since this perturbation may still retain source-pair-specific bias, the second stage performs sparse refinement through attention-shift-filtered multi-objective evolutionary search, leveraging heterogeneous source distributions in transfer settings. By separating universal perturbation learning from sparse evolutionary refinement, MA aims to reduce reliance on source-pair-specific patterns and improve transferability to unseen domains. Extensive experiments demonstrate strong attack effectiveness on cross-modal ReID benchmarks and improved transferability in the evaluated held-out-domain settings, with additional evaluations across model architectures.
Zimeng Wu, Qijian Tian, Penghui Du, Changqian Yu, Jiaxin Chen, Yunhong Wang
Unified Multimodal Models (UMMs) enable powerful multi-image synthesis but incur substantial computation from long visual contexts. While token pruning offers a direct route to accelerating inference, existing methods are primarily adapted from understanding tasks, suffering from insufficient reference image exploration, confused conditional guidance, and pooling-induced ambiguity that degrades generation quality. To address these issues, we propose CATP, the first training-free context-compression method tailored to UMM generation, which \textbf{C}alibrates the discrepancies of general \textbf{A}ttention-based \textbf{T}oken \textbf{P}runing for better quality-efficiency trade-off. Concretely, building on independent layer-wise pruning during denoising, we introduce Guidance-Consistent Pruning (GCP) to synchronize pruning decisions across conditioning branches, and a Taylor-Calibrated Attention Metric (TCAM) that derives a theoretically grounded second-order correction for pooling-induced bias. In addition, a Fast Scoring Operator (FSO) fuses the calibration into tiled online Softmax to reduce pruning cost. Extensive experiments across representative UMMs and benchmarks demonstrate that CATP consistently outperforms state-of-the-art token-reduction methods, achieving up to 2.09$\times$ attention-core speedup.
Chong Wang, Zixuan Fu, Hao Cheng, Siyuan Yang, Alex C. Kot, Bihan Wen
Recent advances in diffusion models have achieved remarkable success in restoring high-quality images in a zero-shot manner. Unfortunately, extending this success to video restoration is challenging due to the substantial computational resources required to pre-train video diffusion models. A rising trend seeks to address video restoration by utilizing a pre-trained image diffusion model, initiating restoration from either frame-wise independent or identical Gaussian noise. Yet, these approaches often fall short, as such simple noise initiations fail to capture and utilize the inherent temporal correlations within degraded video frames. In this work, we propose taming noise initialization for zero-shot video restoration, which preserves the intrinsic temporal correlations by generating initial noises directly using DDIM inversion on the corrupted video frames themselves. Building on this, we introduce an adaptive noise blending strategy that suppresses degradation patterns encoded during DDIM inversion while gradually steering the noise toward a Gaussian distribution. To further emphasize temporal consistency, we propose measurement-guided trajectory attention to better align features across frames. Extensive experiments on multiple video restoration tasks demonstrate that our method outperforms existing zero-shot approaches, consistently achieving superior results on the DAVIS, REDS4, and Set8 datasets.
Flexible image tokenizers aim to represent an image using an ordered 1D variable-length token sequence. This flexible tokenization is typically achieved through nested dropout, where a portion of trailing tokens is randomly truncated during training, and the image is reconstructed using the remaining preceding sequence. However, this tail-truncation strategy inherently concentrates the image information in the early tokens, limiting the effectiveness of downstream AutoRegressive (AR) image generation as the token length increases. To overcome these limitations, we propose \textbf{ReToK}, a flexible tokenizer with \underline{Re}dundant \underline{Tok}en Padding and Hierarchical Semantic Regularization, designed to fully exploit all tokens for enhanced latent modeling. Specifically, we introduce \textbf{Redundant Token Padding} to activate tail tokens more frequently, thereby alleviating information over-concentration in the early tokens. In addition, we apply \textbf{Hierarchical Semantic Regularization} to align the decoding features of earlier tokens with those from a pre-trained vision foundation model, while progressively reducing the regularization strength toward the tail to allow finer low-level detail reconstruction. Extensive experiments demonstrate the effectiveness of ReTok: on ImageNet 256$\times$256, our method achieves superior generation performance compared with both flexible and fixed-length tokenizers. Code will be available at: \href{https://github.com/zfu006/ReTok}{https://github.com/zfu006/ReTok}
Remote sensing object detection is a critical technology for real-world applications such as natural resource monitoring, traffic management, and UAV-based rescue. Detecting tiny objects in high-resolution aerial imagery remains challenging due to weak visual cues and insufficient global context modeling in complex scenes. Existing methods often suffer from delayed contextual interaction and limited nonlinear reasoning, which restrict their ability to effectively refine shallow representations and ultimately lead to suboptimal performance. To address these challenges, we propose FMC-DETR, a frequency-decoupled fusion framework for aerial-view object detection. First, we propose the Wavelet Kolmogorov-Arnold Transformer (WeKat) backbone, which employs cascaded wavelet transforms to enhance global low-frequency structure perception in shallow features while preserving fine-grained details, and further leverages Kolmogorov-Arnold networks for adaptive nonlinear modeling of multi-scale dependencies. Second, we introduce the Multi-Domain Feature Coordination (MDFC) module, which refines cross-scale fused representations through partial-channel spatial, spectral, and structural coordination, thereby strengthening small-object-related feature responses in cluttered scenes. Finally, we design the Compact Partial Fusion (CPF) module, which performs compact multi-branch aggregation with progressive partial refinement to improve feature diversity and multi-scale interaction while preserving stable information flow and reducing redundant perturbation. Extensive experiments across multiple remote sensing benchmarks demonstrate that FMC-DETR achieves state-of-the-art performance and significantly outperforming the baseline detector. Code is available at https://github.com/bloomingvision/FMC-DETR.
With the advancement of vision-language models, web automation has made significant progress. However, deploying autonomous agents in real-world settings remains challenging, primarily due to site heterogeneity, where generalist models lack domain-specific priors for diverse interfaces, and long-horizon instability, characterized by the accumulation of decision drift over extended interactions. To address these challenges, we introduce ColorBrowserAgent (Complex Long-Horizon Browser Agent), a knowledge-evolving agent for robust web automation. Our approach addresses these challenges through two synergistic mechanisms: human-in-the-loop knowledge adaptation that transforms sparse human feedback into reusable domain knowledge, and knowledge-aligned progressive summarization that stabilizes long interactions through memory compression. Extensive experiments on WebArena, WebChoreArena and industrial deployment show that ColorBrowserAgent consistently outperforms strong baselines. It achieves a state-of-the-art success rate of 71.2% on WebArena and maintains 47.4% performance under zero-shot transfer setting on WebChoreArena. In commercial deployment, it improves user satisfaction by 19.3% relatively, verifying its robustness in real-world scenarios.
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 6: Industry Track) · 2026 · 正式发表