Real-world visual data rarely presents as isolated, static instances. Instead, it often evolves gradually over time through variations in pose, lighting, object state, or scene context. However, conventional classifiers are typically trained under the assumption of temporal independence, limiting their ability to capture such dynamics. We propose a simple yet effective framework that equips standard feedforward classifiers with temporal reasoning, all without modifying model architectures or introducing recurrent modules. At the heart of our approach is a novel Support-Exemplar-Query (SEQ) learning paradigm, which structures training data into temporally coherent trajectories. These trajectories enable the model to learn class-specific temporal prototypes and align prediction sequences via a differentiable soft-DTW loss. A multi-term objective further promotes semantic consistency and temporal smoothness. By interpreting input sequences as evolving feature trajectories, our method introduces a strong temporal inductive bias through loss design alone. This proves highly effective in both static and temporal tasks: it enhances performance on fine-grained and ultra-fine-grained image classification, and delivers precise, temporally consistent predictions in video anomaly detection. Despite its simplicity, our approach bridges static and temporal learning in a modular and data-efficient manner, requiring only a simple classifier on top of pre-extracted features.
论文检索
输入标题、作者或关键词，从 1,526 篇学术成果中精准定位
Recently, Omni-modal large language models (OLLMs) have sparked a new wave of research, achieving impressive results in tasks such as audio-video understanding and real-time environment perception. However, hallucination issues still persist. Similar to the bimodal setting, the priors from the text modality tend to dominate, leading OLLMs to rely more heavily on textual cues while neglecting visual and audio information. In addition, fully multimodal scenarios introduce new challenges. Most existing models align visual or auditory modalities with text independently during training, while ignoring the intrinsic correlations between video and its corresponding audio. This oversight results in hallucinations when reasoning requires interpreting hidden audio cues embedded in video content. To address these challenges, we propose OmniDPO, a preference-alignment framework designed to mitigate hallucinations in OLLMs. Specifically, OmniDPO incorporates two strategies: (1) constructing text-preference sample pairs to enhance the model’s understanding of audio-video interactions; and (2) constructing multimodal-preference sample pairs to strengthen the model’s attention to visual and auditory information. By tackling both challenges, OmniDPO effectively improves multimodal grounding and reduces hallucination. Experiments conducted on two OLLMs demonstrate that OmniDPO not only effectively mitigates multimodal hallucinations but also significantly enhances the models' reasoning capabilities across modalities.
Medical Large Vision-Language Models (Med-LVLMs) have shown promising results in clinical applications, but often suffer from hallucinated outputs due to misaligned visual understanding. In this work, we identify two fundamental limitations contributing to this issue: insufficient visual representation learning and poor visual attention alignment. To address these problems, we propose MedAlign, a simple, lightweight alignment distillation framework that transfers visual alignment knowledge from a domain-specific Contrastive Language-Image Pre-training (CLIP) model to Med-LVLMs. MedAlign introduces two distillation losses: a spatial-aware visual alignment loss based on visual token-level similarity structures, and an attention-aware distillation loss that guides attention toward diagnostically relevant regions. Extensive experiments on medical report generation and medical visual question answering (VQA) benchmarks show that MedAlign consistently improves both performance and interpretability, yielding more visually grounded outputs.
Text-to-image diffusion models utilize cross-attention to integrate textual information into the visual latent space, yet the transformation from text embeddings to latent features remains largely unexplored. We provide a mechanistic analysis of the output-value (OV) circuits within cross-attention layers through spectral analysis via singular value decomposition. Our analysis reveals that semantic concepts are encoded in low-dimensional subspaces spanned by singular vectors in OV circuits across cross-attention heads. To verify this, we intervene on concept-related components in the diffusion process, demonstrating that intervention on identified spectral components affects conceptual changes. We further validate these findings by examining visual outputs of isolated subspaces and their alignment with text embedding space. Through this mechanistic understanding, we demonstrate that only nullifying these spectral components can achieve targeted concept removal with performance comparable to existing methods while providing interpretability. Our work reveals how cross-attention layers encode semantic concepts in spectral subspaces of OV circuits, providing mechanistic insights and enabling precise concept manipulation without retraining.
Robot navigation in dense crowds requires understanding social cues that humans naturally use, yet existing methods struggle with real-world complexity. We investigate two questions: (1) Where do pedestrians look when navigating crowds? and (2) Can eye tracking improve robot navigation? To answer, we introduce GazeNav, an egocentric dataset collected via wearable eye trackers, featuring synchronized video, gaze, and trajectories in crowded environments. Analysis reveals that the gaze of pedestrians is closely related to the semantic presence and movement of other individuals, exhibiting distinct attention patterns across navigation behaviors. Building on this, we propose Gaze2Nav, a modular framework that first predicts human gaze to infer socially salient pedestrians, then incorporates the semantic attention into motion planning alongside visual inputs. Our method achieves 87.6% salient pedestrian prediction accuracy and reduces trajectory error by 15.4% over state-of-the-art baselines. By aligning with human gaze, our framework improves both performance and interpretability, advancing toward human-like, socially intelligent robot navigation.
Traditional vision-based material perception methods often experience substantial performance degradation under visually impaired conditions, thereby motivating the shift toward non-visual multimodal material perception. Despite this, existing approaches frequently perform naive fusion of multimodal inputs, overlooking key challenges such as modality-specific noise, missing modalities common in real-world scenarios, and the dynamically varying importance of each modality depending on the task. These limitations lead to suboptimal performance across several benchmark tasks. In this paper, we propose a robust multimodal fusion framework, TouchFormer. Specifically, we employ a Modality-Adaptive Gating (MAG) mechanism and intra- and inter-modality attention mechanisms to adaptively integrate cross-modal features, enhancing model robustness. Additionally, we introduce a Cross-Instance Embedding Regularization(CER) strategy, which significantly improves classification accuracy in fine-grained subcategory material recognition tasks. Experimental results demonstrate that, compared to existing non-visual methods, the proposed TouchFormer framework achieves classification accuracy improvements of 2.48% and 6.83% on SSMC and USMC tasks, respectively. Furthermore, real-world robotic experiments validate TouchFormer's effectiveness in enabling robots to better perceive and interpret their environment, paving the way for its deployment in safety-critical applications such as emergency response and industrial automation.
SemanticVLA: Semantic-Aligned Sparsification and Enhancement for Efficient Robotic Manipulation
PDF ↗Vision-Language-Action (VLA) models have advanced in robotic manipulation, yet practical deployment remains hindered by two key limitations: **1) perceptual redundancy**, where irrelevant visual inputs are processed inefficiently, and **2) superficial instruction-vision alignment**, which hampers semantic grounding of actions. In this paper, we propose **SemanticVLA**, a novel VLA framework that performs Semantic-Aligned Sparsification and Enhancement for Efficient Robotic Manipulation. Specifically: **1)** To sparsify redundant perception while preserving semantic alignment, **Semantic-guided Dual Visual Pruner (SD-Pruner)** performs: Instruction-driven Pruner (ID-Pruner) extracts global action cues and local semantic anchors in SigLIP; Spatial-aggregation Pruner (SA-Pruner) compacts geometry-rich features into task-adaptive tokens in DINOv2. **2)** To exploit sparsified features and integrate semantics with spatial geometry, **Semantic-complementary Hierarchical Fuser (SH-Fuser)** fuses dense patches and sparse tokens across SigLIP and DINOv2 for coherent representation. **3)** To enhance the transformation from perception to action, **Semantic-conditioned Action Coupler (SA-Coupler)** replaces the conventional observation-to-DoF approach, yielding more efficient and interpretable behavior modeling for manipulation tasks. Extensive experiments on simulation and real-world tasks show that SemanticVLA sets a new SOTA in both performance and efficiency. SemanticVLA surpasses OpenVLA on LIBERO benchmark by **21.1%** in success rate, while reducing training cost and inference latency by **3.0×** and **2.7×**.
Diffusion policies have recently shown great promise for generating actions in robotic manipulation. However, existing approaches often rely on global instructions to produce short-term control signals, which can result in misalignment in action generation. We conjecture that the primitive skills, referred to as fine-grained, short-horizon manipulations, such as "move up" and "open the gripper", provide a more intuitive and effective interface for robot learning. To bridge this gap, we propose SDP, a skill-conditioned diffusion policy that integrates interpretable skill learning with conditional action planning. SDP abstracts eight reusable primitive skills across tasks and employs a vision-language model to extract discrete representations from visual observations and language instructions. Based on the representations, a lightweight router network is designed to assign a desired primitive skill for each state, which helps construct a single-skill policy to generate skill-aligned actions. By decomposing complex tasks into a sequence of primitive skills and selecting a single-skill policy, the proposed SDP ensures skill-consistent behavior across diverse tasks. Extensive experiments on two challenging simulation benchmarks and real-world robot deployments demonstrate that SDP consistently outperforms state-of-the-art methods, providing a new paradigm for skill-based robot learning with diffusion policies.
While natural language is commonly used to guide embodied agents, the inherent ambiguity and verbosity of language often hinder the effectiveness of language-guided navigation in complex environments. To this end, we propose Visual Prompt Navigation (VPN), a novel paradigm that guides agents to navigate using only user-provided visual prompts within 2D top-view maps. This visual prompt primarily focuses on marking the visual navigation trajectory on a top-down view of a scene, offering intuitive and spatially grounded guidance without relying on language instructions. It is more friendly for non-expert users and reduces interpretive ambiguity. We build VPN tasks in both discrete and continuous navigation settings, constructing two new datasets, R2R-VP and R2R-CE-VP, by extending existing R2R and R2R-CE episodes with corresponding visual prompts. Furthermore, we introduce VPNet, a dedicated baseline network to handle the VPN tasks, with two data augmentation strategies: view-level augmentation (altering initial headings and prompt orientations) and trajectory-level augmentation (incorporating diverse trajectories from large-scale 3D scenes), to enhance navigation performance. Extensive experiments evaluate how visual prompt forms, top-view map formats, and data augmentation strategies affect the performance of visual prompt navigation.
Plug-and-Play Clarifier: A Zero-Shot Multimodal Framework for Egocentric Intent Disambiguation
PDF ↗The performance of egocentric AI agents is fundamentally limited by multimodal intent ambiguity. This challenge arises from a combination of underspecified language, imperfect visual data, and deictic gestures, which frequently leads to task failure. Existing monolithic Vision-Language Models (VLMs) struggle to resolve these multimodal ambiguous inputs, often failing silently or hallucinating responses. To address these ambiguities, we introduce the Plug-and-Play Clarifier, a zero-shot and modular framework that decomposes the problem into discrete, solvable sub-tasks. Specifically, our framework consists of three synergistic modules: (1) a text clarifier that uses dialogue-driven reasoning to interactively disambiguate linguistic intent, (2) a vision clarifier that delivers real-time guidance feedback, instructing users to adjust their positioning for improved capture quality, and (3) a cross-modal clarifier with grounding mechanism that robustly interprets 3D pointing gestures and identifies the specific objects users are pointing to. Extensive experiments demonstrate that our framework improves the intent clarification performance of small language models (4-8B) by approximately 30%, making them competitive with significantly larger counterparts. We also observe consistent gains when applying our framework to these larger models. Furthermore, our vision clarifier increases corrective guidance accuracy by over 20%, and our cross-modal clarifier improves semantic answer accuracy for referential grounding by 5%. Overall, our method provides a plug-and-play framework that effectively resolves multimodal ambiguity and significantly enhances user experience in egocentric interaction.
Existing evaluation protocols for brain visual decoding predominantly rely on coarse metrics that obscure inter-model differences, lack neuroscientific foundation, and fail to capture fine-grained visual distinctions. To address these limitations, we introduce BASIC, a unified, multigranular evaluation framework that jointly quantifies structural fidelity, inferential alignment, and contextual coherence between decoded and ground-truth images. For the structural level, we introduce a hierarchical suite of segmentation-based metrics, including foreground, semantic, instance, and component masks, anchored in granularity-aware correspondence across mask structures. For the semantic level, we extract structured scene representations encompassing objects, attributes, and relationships using multimodal large language models, enabling detailed, scalable, and context-rich comparisons with ground-truth stimuli. We benchmark a diverse set of visual decoding methods across multiple stimulus-neuroimaging datasets within this unified evaluation framework. Together, these criteria provide a more discriminative, interpretable, and comprehensive foundation for evaluating brain visual decoding methods.
QueryCraft: Transformer-Guided Query Initialization for Enhanced Human-Object Interaction Detection
PDF ↗Human-Object Interaction (HOI) detection aims to localize human-object pairs and recognize their interactions in images. Although DETR-based methods have recently emerged as the mainstream framework for HOI detection, they still suffer from a key limitation: Randomly initialized queries lack explicit semantics, leading to suboptimal detection performance. To address this challenge, we propose QueryCraft, a novel plug-and-play HOI detection framework that incorporates semantic priors and guided feature learning through transformer-based query initialization. Central to our approach is ACTOR (Action-aware Cross-modal TransfORmer), a cross-modal Transformer encoder that jointly attends to visual regions and textual prompts to extract action-relevant features. Rather than merely aligning modalities, ACTOR leverages language-guided attention to infer interaction semantics and produce semantically meaningful query representations. To further enhance object-level query quality, we introduce a Perceptual Distilled Query Decoder (PDQD), which distills object category awareness from a pre-trained detector to serve as object query initiation. This dual-branch query initialization enables the model to generate more interpretable and effective queries for HOI detection. Extensive experiments on HICO-Det and V-COCO benchmarks demonstrate that our method achieves state-of-the-art performance and strong generalization.
A Brain-Inspired Saliency Prediction Framework for Human-AI Cognitive Consistency in AIGC Content via Multi-Region Liquid Neurons
PDF ↗In recent years, human-AI cognitive consistency has emerged as a crucial perspective for evaluating the perceptual quality and interpretability of AIGC (Artificial Intelligence Generated Content). This paper proposes a biologically inspired saliency prediction framework that models six core regions of the human visual system—namely V1, V2, V4, MT, LIP, and FEF—using liquid neurons to capture the dynamic saliency features aligned with human gaze behavior. To enable effective alignment between AIGC models and human cognitive mechanisms, we introduce a cross-domain dual-teacher distillation strategy and construct a large-scale multimodal dataset comprising natural images, eye-tracking data, AIGC-generated images, and their corresponding cross-attention maps. Furthermore, we propose HAMCI (Human-AI Mutual Cognitive Index), a novel metric designed to quantitatively assess the spatial and semantic alignment between predicted saliency maps and model attention distributions. The proposed method demonstrates promising performance across various saliency prediction and cognitive alignment tasks, with results comparable to or surpassing recent state-of-the-art methods in several benchmarks. The code and dataset will be released upon acceptance to facilitate future research on cognitively aligned AIGC evaluation.
PINet: Improving the Stability of Prototype Networks via Phantasia-Inspired Uncertain Representations
PDF ↗Self-interpretable models are increasingly valued for their inherent explainability. Among them, part-prototype networks stand out by mimicking human reasoning through the use of learned prototypes. However, their explanations often lack stability, becoming sensitive to subtle input perturbations. In this work, we propose Prototype in Imagery Network (PINet), a framework that improves the stability of prototype-based explanations. Rather than training on all possible input variations, which is computationally infeasible, PINet draws inspiration from visual mental imagery. Specifically, we incorporate empty inputs and apply coarse location guidance to simulate the human ability to imagine rough object features (a process akin to Phantasia). PINet mimics this process by incorporating empty inputs and applying coarse location guidance. These imagined, or uncertain, representations are contrasted with those derived from actual inputs (certain representations). We model the differences between the two by computing similarity at both the feature and prototype levels, allowing uncertainty to be explicitly encoded during prototype learning. Comprehensive evaluations on CUB-200-2011 and Stanford Cars demonstrate that PINet consistently achieves robust accuracy and localization, even under noisy conditions. These results represent the ability of PINet to produce stable and interpretable explanations under uncertainty.
Do Large Language Models Reason About Uncertainty Like Humans? A Benchmark on Hurricane Forecast Visualization Comprehension
PDF ↗Uncertainty visualizations, such as hurricane cones and ensemble tracks, are essential for risk communication but are often misinterpreted, leading to harmful decisions. As AI assistants like large language models (LLMs) increasingly support understanding of graphics and decision-making, they offer a promising pathway to enhance the interpretation of complex visualizations and a new opportunity to examine and improve the interpretation of uncertainty. We introduce UnReason, the first benchmark that systematically compares how humans and LLMs reason about hurricane forecast uncertainty visualizations. UnReason spans two escalating phases, seven representative visualization formats, six real hurricane cases, and three agent types (humans, LLMs with context, and LLMs without context), including 880 visualizations and 117,600 structured question–answer pairs under matched evaluation conditions. Phase 1 evaluates reasoning across implicit and explicit uncertainty encodings; Phase 2 examines reasoning under single- versus multi-dimensional uncertainty representations. We thoroughly assess damage estimation, reasoning strategies, and comprehension patterns, revealing that LLMs have a stronger semantic and conceptual understanding of uncertainty, and are less misled by visual variability, but still replicate key human biases during decision-making. Our findings offer insights into aligning LLM behavior with human cognition in uncertainty-rich visual reasoning tasks.
Explainable AI (XAI) methods generally fall into two categories. Post-hoc approaches generate explanations for pre-trained models and are compatible with various neural network architectures. These methods often use feature importance visualizations, such as saliency maps, to indicate which input regions influenced the model’s prediction. Unfortunately, they typically offer a coarse understanding of the model’s decision-making process. In contrast, ante-hoc (inherently explainable) methods rely on specially designed model architectures trained from scratch. A notable subclass of these methods provides explanations through prototypes, representative patches extracted from the training data. However, prototype-based approaches require dedicated architectures, involve specialized training procedures, and perform well only on specific datasets. In this work, we propose EPIC (Explanation of Pretrained Image Classification), a novel approach that bridges the gap between these two paradigms. Like post-hoc methods, EPIC operates on pre-trained models without architectural modifications. Simultaneously, it delivers intuitive, prototype-based explanations inspired by ante-hoc techniques. To the best of our knowledge, EPIC is the first post-hoc method capable of fully replicating the core explanatory power of inherently interpretable models. We evaluate EPIC on benchmark datasets commonly used in prototype-based explanations, such as CUB-200-2011 and Stanford Cars, alongside large-scale datasets like ImageNet, typically employed by post-hoc methods. EPIC uses prototypes to explain model decisions, providing a flexible and easy-to-understand tool for creating clear, high-quality explanations.
Vision-language models bridge visual and linguistic understanding and have proven to be powerful for video recognition tasks. Existing methods primarily rely on parameter-efficient fine-tuning of pre-trained image-text models, suffering from limited interpretability and poor generalization due to inadequate temporal modeling. To address these, we propose a simple yet effective video-to-text discretization framework. Our approach leverages the frozen text encoder to build a visual codebook derived from video class labels, exploiting the many-to-one contrastive alignment between visual and textual embeddings in multimodal pretraining. This enables the transformation of temporal visual features into discrete textual tokens via feature lookups, yielding interpretable video representations through explicit video modeling. Then, to improve robustness against noisy or irrelevant frames, we introduce a confidence-aware fusion module that dynamically weights keyframes based on their semantic relevance, as measured by the codebook. Furthermore, we incorporate learnable text prompts to conduct adaptive codebook updates during training. Experiments on four datasets, including HMDB-51, UCF-101, Something-Something-v2, and Kinetics-400, validate the superiority of our approach, achieving competitive improvements over state-of-the-art approaches.
Accurate medical diagnosis often involves progressive visual focusing and iterative reasoning, characteristics commonly observed in clinical workflows. While recent vision-language models demonstrate promising chain-of-thought (CoT) reasoning capabilities via reinforcement learning with verifiable rewards (RLVR), their purely on-policy learning paradigm tends to reinforce superficially coherent but clinically inaccurate reasoning paths. We propose MedEyes, a novel reinforcement learning framework that dynamically models clinician-style diagnostic reasoning by progressively attending to and interpreting relevant medical image regions. By incorporating off-policy expert guidance, MedEyes converts expert visual search trajectories into structured external behavioral signals, guiding the model toward clinically aligned visual reasoning. We design the Gaze-guided Reasoning Navigator (GRN) to emulate the diagnostic process through a dual-mode exploration strategy, scanning for systematic abnormality localization and drilling for detailed regional analysis. To balance expert imitation and autonomous discovery, we introduce the Confidence Value Sampler (CVS), which employs nucleus sampling and adaptive termination to create diverse yet credible exploration paths. Finally, the dual-stream GRPO optimization framework decouples on-policy and off-policy learning signals, mitigating reward assimilation and entropy collapse. Experiments demonstrate that MedEyes achieves an average performance improvement of +8.5% across multiple medical VQA benchmarks, validating MedEyes's potential in building trustworthy medical AI systems.
Despite the rapid progress of Vision Language Models (VLMs), existing benchmarks still concentrate on coarse-grained object recognition or simple relational reasoning, leaving the fine-grained and higher-order reasoning abilities of these systems largely unexamined. To bridge this critical evaluation gap, we introduce EmojiGrid, a novel diagnostic benchmark specifically designed to probe these fine-grained and higher-order skills. Leveraging the universal and semantically rich nature of emojis, we synthesize a grid‑based visual dataset paired with 29,000+ QA pairs. Each pair is explicitly anchored in a three-level cognitive taxonomy comprising (i) Perception and Information Extraction, (ii) Relational and Structural Reasoning, and (iii) Abstraction and Advanced Cognition. These dimensions further decompose into nine categories covering a broad range of cognitive skills, including counting, spatial relations, compositional logic, semantic sentiment, and related higher-order reasoning tasks. Our extensive evaluation of 25 state-of-the-art open-source and proprietary VLMs reveals a significant performance gap between foundational perceptual tasks and higher-level cognitive abilities, particularly in abstraction and advanced emotional reasoning. Notably, all models struggle with compositional logic, spatial consistency, and especially emotional and semantic understanding. EmojiGrid provides a quantifiable, fine-grained benchmark to diagnose VLM limitations and guides future progress toward models that can truly perceive, reason about, and interpret complex, symbol-rich visual scenes.
Understanding anomalous human behaviors at a fine-grained level remains a major challenge in complex scenarios. Existing video anomaly understanding (VAU) methods often rely on coarse frame-level cues or overlook structured modeling of individual actions, limiting their capacity for reasoning about human interactions and accountability. To address these challenges, we propose TargetVAU, a multimodal anomaly-aware reasoning framework designed for individual-level anomaly recognition and explanation. TargetVAU first extracts both global-level and human-centric visual features using a frozen Vision Transformer (ViT) encoder. An Anomaly-focused Temporal Sampler is then employed to select behaviorally informative frames via a density-aware strategy guided by predicted anomaly scores. A Spatio-Temporal Interaction Graph is constructed to explicitly model interactions among individuals across time and space. These structured representations are fused with prompt embeddings via a frozen Q-Former to form a unified semantic representation. Finally, a large language model fine-tuned with low-rank adaptation (LoRA) performs instruction-guided reasoning to identify anomalous individuals and generate natural language explanations. Extensive experiments on UCCD and HIVAU-70K demonstrate that TargetVAU significantly outperforms existing methods in both accuracy and interpretability, advancing the state of individual-level anomaly understanding in surveillance videos.