Fine-Grained Visual Perception

Retrieval-Augmented Generation (RAG) extends Large Vision-Language Models (LVLMs) with external visual knowledge. However, existing visual RAG systems typically rely on generic retrieval signals that overlook the fine-grained visual semantics essential for complex reasoning. To address this limitation, we propose UniDoc-RL, a unified reinforcement learning framework in which an LVLM agent jointly performs retrieval, reranking, active visual perception, and reasoning. UniDoc-RL formulates visual information acquisition as a sequential decision-making problem with a hierarchical action space. Specifically, it progressively refines visual evidence from coarse-grained document retrieval to fine-grained image selection and active region cropping, allowing the model to suppress irrelevant content and attend to information-dense regions. For effective end-to-end training, we introduce a dense multi-reward scheme that provides task-aware supervision for each action. Based on Group Relative Policy Optimization (GRPO), UniDoc-RL aligns agent behavior with multiple objectives without relying on a separate value network. To support this training paradigm, we curate a comprehensive dataset of high-quality reasoning trajectories with fine-grained action annotations. Experiments on three benchmarks demonstrate that UniDoc-RL consistently surpasses state-of-the-art baselines, yielding up to 17.7% gains over prior RL-based methods.

Few-shot fine-grained image classification aims to recognize subcategories with high visual similarity using only a limited number of annotated samples. Existing metric learning-based methods typically rely solely on spatial domain features. Confined to this single perspective, models inevitably suffer from inherent texture biases, entangling essential structural details with high-frequency background noise. Furthermore, lacking cross-view geometric constraints, single-view metrics tend to overfit this noise, resulting in structural instability under few-shot conditions. To address these issues, this paper proposes the Frequency-Enhanced Dual-Subspace Network (FEDSNet). Specifically, FEDSNet utilizes the Discrete Cosine Transform (DCT) and a low-pass filtering mechanism to explicitly isolate low-frequency global structural components from spatial features, thereby suppressing background interference. Truncated Singular Value Decomposition (SVD) is employed to construct independent, low-rank linear subspaces for both spatial texture and frequency structural features. An adaptive gating mechanism is designed to dynamically fuse the projection distances from these dual views. This strategy leverages the structural stability of the frequency subspace to prevent the spatial subspace from overfitting to background features. Extensive experiments on four benchmark datasets - CUB-200-2011, Stanford Cars, Stanford Dogs, and FGVC-Aircraft - demonstrate that FEDSNet exhibits excellent classification performance and robustness, achieving highly competitive results compared to existing metric learning algorithms. Complexity analysis further confirms that the proposed network achieves a favorable balance between high accuracy and computational efficiency, providing an effective new paradigm for few-shot fine-grained visual recognition.

While Multimodal Large Language Models (MLLMs) have demonstrated impressive capabilities in general visual understanding, they frequently falter in fine-grained perception tasks that require identifying tiny objects or discerning subtle visual relationships. We attribute this limitation to Visual Attenuation: a phenomenon where sparse fine-grained visual signals are prematurely suppressed or diluted by dominant textual tokens during network propagation, resulting in a "loss of focus" during the deep-level decision-making process. Existing input-centric solutions fail to fundamentally reverse this intrinsic mechanism of information loss. To address this challenge, we propose the Variational Information Flow (VIF) framework. Adopting a probabilistic perspective, VIF leverages a Conditional Variational Autoencoder (CVAE) to model the visual saliency relevant to the question-answer pair as a latent distribution. As a plug-and-play module, VIF can be integrated into existing architectures. Extensive evaluations across diverse benchmarks, covering General VQA, fine-grained perception, and visual grounding, demonstrate that VIF yields competitive improvements over previous methods, validating its effectiveness in enhancing the fine-grained perception of MLLMs.

Recent multimodal large language models (MLLMs) support Thinking with Images, invoking visual tools such as zooming and cropping to inspect image regions during inference. Yet these systems remain brittle in fine-grained reasoning: to acquire a decisive detail, a model must ground its attention on the correct region, but knowing which region is correct presupposes having already observed that detail. We identify this circular dependency as the grounding paradox, show that grounding errors are rarely self-corrected within a single trajectory---once a misleading region is inspected, all subsequent reasoning conditions on that observation and the error propagates to the final answer---and observe that because each trajectory constructs its own evidence, answer-level aggregation discards the very information that distinguishes trajectories. We propose Test-Time Scaling over Perception (TTSP), a closed-loop framework that treats perception as the unit of scalable inference and allocates compute along two axes: Entropy-Gated Perceptual Exploration samples diverse trajectories and uses critical-token entropy to withhold evidence the model cannot commit to, while Evidence-Guided Iterative Refinement distills validated observations into a correctable Evidence Ledger that steers later rounds to re-inspect unresolved regions. Across high-resolution and general multimodal benchmarks, TTSP consistently outperforms strong test-time scaling baselines, while improving grounding quality with favorable token efficiency.

Accurate assessment of spheno-occipital synchondrosis (SOS) maturation is a key indicator of craniofacial growth and a critical determinant for orthodontic and surgical timing. However, SOS staging from cone-beam CT (CBCT) relies on subtle, continuously evolving morphological cues, leading to high inter-observer variability and poor reproducibility, especially at transitional fusion stages. We frame SOS assessment as a fine-grained visual recognition problem and propose a progressive representation-learning framework that explicitly mirrors how expert clinicians reason about synchondral fusion: from coarse anatomical structure to increasingly subtle patterns of closure. Rather than training a full-capacity network end-to-end, we sequentially grow the model by activating deeper blocks over time, allowing early layers to first encode stable cranial base morphology before higher-level layers specialize in discriminating adjacent maturation stages. This yields a curriculum over network depth that aligns deep feature learning with the biological continuum of SOS fusion. Extensive experiments across convolutional and transformer-based architectures show that this expert-inspired training strategy produces more stable optimization and consistently higher accuracy than standard training, particularly for ambiguous intermediate stages. Importantly, these gains are achieved without changing network architectures or loss functions, demonstrating that training dynamics alone can substantially improve anatomical representation learning. The proposed framework establishes a principled link between expert dental intuition and deep visual representations, enabling robust, data-efficient SOS staging from CBCT and offering a general strategy for modeling other continuous biological processes in medical imaging.

Recent advancements in Vision-Language Models (VLMs) have revolutionized general visual understanding. However, their application in the food domain remains constrained by benchmarks that rely on coarse-grained categories, single-view imagery, and inaccurate metadata. To bridge this gap, we introduce DiningBench, a hierarchical, multi-view benchmark designed to evaluate VLMs across three levels of cognitive complexity: Fine-Grained Classification, Nutrition Estimation, and Visual Question Answering. Unlike previous datasets, DiningBench comprises 3,021 distinct dishes with an average of 5.27 images per entry, incorporating fine-grained "hard" negatives from identical menus and rigorous, verification-based nutritional data. We conduct an extensive evaluation of 29 state-of-the-art open-source and proprietary models. Our experiments reveal that while current VLMs excel at general reasoning, they struggle significantly with fine-grained visual discrimination and precise nutritional reasoning. Furthermore, we systematically investigate the impact of multi-view inputs and Chain-of-Thought reasoning, identifying five primary failure modes. DiningBench serves as a challenging testbed to drive the next generation of food-centric VLM research. All codes are released in https://github.com/meituan/DiningBench.

Extracting vehicle information from surveillance images is essential for intelligent transportation systems, enabling applications such as traffic monitoring and criminal investigations. While Automatic License Plate Recognition (ALPR) is widely used, Fine-Grained Vehicle Classification (FGVC) offers a complementary approach by identifying vehicles based on attributes such as color, make, model, and type. Although there have been advances in this field, existing studies often assume well-controlled conditions, explore limited attributes, and overlook FGVC integration with ALPR. To address these gaps, we introduce UFPR-VeSV, a dataset comprising 24,945 images of 16,297 unique vehicles with annotations for 13 colors, 26 makes, 136 models, and 14 types. Collected from the Military Police of Paraná (Brazil) surveillance system, the dataset captures diverse real-world conditions, including partial occlusions, nighttime infrared imaging, and varying lighting. All FGVC annotations were validated using license plate information, with text and corner annotations also being provided. A qualitative and quantitative comparison with established datasets confirmed the challenging nature of our dataset. A benchmark using five deep learning models further validated this, revealing specific challenges such as handling multicolored vehicles, infrared images, and distinguishing between vehicle models that share a common platform. Additionally, we apply two optical character recognition models to license plate recognition and explore the joint use of FGVC and ALPR. The results highlight the potential of integrating these complementary tasks for real-world applications. The UFPR-VeSV dataset is publicly available at: https://github.com/Lima001/UFPR-VeSV-Dataset.

Vision-Language Models (VLMs) excel on many multimodal reasoning benchmarks, but these evaluations often do not require an exhaustive readout of the image and can therefore obscure failures in faithfully capturing all visual details. We introduce Grid2Matrix (G2M), a controlled benchmark in which a model is shown a color grid and a color-to-number mapping, and must output the corresponding matrix. By varying grid size and the number of colors, G2M provides a simple way to increase visual complexity while minimizing semantic confounds. We find that VLMs exhibit a sharp early collapse in zero-shot end-to-end evaluation, failing on surprisingly small grids rather than degrading gradually as the task becomes denser. We probe the visual encoders of VLMs from two representative families and find that they preserve substantially more of the grid information than the corresponding end-to-end outputs. This suggests that the failure is not explained by visual encoding alone, but also reflects a gap between what remains recoverable from visual features and what is ultimately expressed in language. We term this gap Digital Agnosia. Further analyses show that these errors are highly structured and depend strongly on how grid cells overlap with visual patch boundaries. We also find that common strategies such as model scaling and multimodal alignment do not fully eliminate this failure mode. We expect G2M to serve as a useful testbed for understanding where and how VLMs lose fine visual details, and for evaluating tasks where missing even small visual details can matter, such as tables, charts, forms, and GUIs.

Vision-language models (VLMs) have achieved impressive performance across a wide range of multimodal tasks. However, they often fail on tasks that require fine-grained visual perception, even when the required information is still present in their internal representations. Prior work has attributed this ``hidden-in-plain-sight'' gap to the language model, but the cause remains unexplained. In this work, we demonstrate that this gap arises from the language model's lack of semantic labels for fine-grained visual details: when visual entities can be mapped to known concepts, VLMs bypass visual comparison and reason through language; when they cannot, VLMs resort to brittle and hallucinated descriptions. We verify this across semantic correspondence, synthetic shape matching, and face matching, and find that VLMs perform much better when the relevant entities are nameable than when they are unnamable. Mechanistically, Logit Lens analysis confirms that VLMs explicitly recover semantic labels for nameable entities and surface more unique tokens compared to unnameable entities. Furthermore, we show that this limitation can be addressed: teaching completely arbitrary names for unknown entities improves performance. More importantly, task-specific finetuning yields even stronger generalization without relying on language priors, i.e., through real visual perception. Our findings suggest that current VLM failures on visual tasks reflect a learned shortcut rather than a fundamental limitation of multimodal reasoning. Code and datasets are available at https://github.com/Patchwork53/VLMs-Need-Words-COLM2026.

Multimodal Large Language Models (MLLMs) have achieved remarkable success, yet they remain prone to perception-related hallucinations in fine-grained tasks. This vulnerability arises from a fundamental limitation: their reasoning is largely restricted to the language domain, treating visual input as a static, reasoning-agnostic preamble rather than a dynamic participant. Consequently, current models act as passive observers, unable to re-examine visual details to ground their evolving reasoning states. To overcome this, we propose V-Reflection, a framework that transforms the MLLM into an active interrogator through a "think-then-look" visual reflection mechanism. During reasoning, latent states function as dynamic probes that actively interrogate the visual feature space, grounding each reasoning step for task-critical evidence. Our approach employs a two-stage distillation strategy. First, the Box-Guided Compression Module (BCM) establishes stable pixel-to-latent targets through explicit spatial grounding. Next, a Dynamic Autoregressive Compression (DAC) module maps the model's hidden states into dynamic probes that interrogate the global visual feature map. By distilling the spatial expertise of the BCM teacher into the DAC student, V-Reflection internalizes the ability to localize task-critical evidence. During inference, both modules remain entirely inactive, maintaining a purely end-to-end autoregressive decoding in the latent space with optimal efficiency. Extensive experiments demonstrate the effectiveness of our V-Reflection across six perception-intensive benchmarks, significantly narrowing the fine-grained perception gap. Visualizations confirm that latent reasoning autonomously localizes task-critical visual evidence.

To enhance the perception and reasoning capabilities of multimodal large language models in complex visual scenes, recent research has introduced agent-based workflows. In these works, MLLMs autonomously utilize image cropping tool to analyze regions of interest for question answering. While existing training strategies, such as those employing supervised fine-tuning and reinforcement learning, have made significant progress, our empirical analysis reveals a key limitation. We demonstrate the model's strong reliance on global input and its weak dependence on the details within the cropped region. To address this issue, we propose a novel two-stage reinforcement learning framework that does not require trajectory supervision. In the first stage, we introduce the ``Information Gap" mechanism by adjusting the granularity of the global image. This mechanism trains the model to answer questions by focusing on cropped key regions, driven by the information gain these regions provide. The second stage further enhances cropping precision by incorporating a grounding loss, using a small number of bounding box annotations. Experiments show that our method significantly enhances the model's attention to cropped regions, enabling it to achieve state-of-the-art performance on high-resolution visual question-answering benchmarks. Our method provides a more efficient approach for perceiving and reasoning fine-grained details in MLLMs. Code is available at: https://github.com/XuanPu-Z/LFPC.

Fine-grained open-vocabulary object detection (FG-OVD) aims to detect novel object categories described by attribute-rich texts. While existing open-vocabulary detectors show promise at the base-category level, they underperform in fine-grained settings due to the semantic entanglement of subjects and attributes in pretrained vision-language model (VLM) embeddings -- leading to over-representation of attributes, mislocalization, and semantic drift in embedding space. We propose GUIDED, a decomposition framework specifically designed to address the semantic entanglement between subjects and attributes in fine-grained prompts. By separating object localization and fine-grained recognition into distinct pathways, HUIDED aligns each subtask with the module best suited for its respective roles. Specifically, given a fine-grained class name, we first use a language model to extract a coarse-grained subject and its descriptive attributes. Then the detector is guided solely by the subject embedding, ensuring stable localization unaffected by irrelevant or overrepresented attributes. To selectively retain helpful attributes, we introduce an attribute embedding fusion module that incorporates attribute information into detection queries in an attention-based manner. This mitigates over-representation while preserving discriminative power. Finally, a region-level attribute discrimination module compares each detected region against full fine-grained class names using a refined vision-language model with a projection head for improved alignment. Extensive experiments on FG-OVD and 3F-OVD benchmarks show that GUIDED achieves new state-of-the-art results, demonstrating the benefits of disentangled modeling and modular optimization. Our code will be released at https://github.com/lijm48/GUIDED.

Recent advances in Large Vision-Language Models (LVLMs) have enabled training-free Fine-Grained Visual Recognition (FGVR). However, effectively exploiting LVLMs for FGVR remains challenging due to the inherent visual ambiguity of subordinate-level categories. Existing methods predominantly adopt either retrieval-oriented or reasoning-oriented paradigms to tackle this challenge, but both are constrained by two fundamental limitations:(1) They apply the same inference pipeline to all samples without accounting for uneven recognition difficulty, thereby leading to suboptimal accuracy and efficiency; (2) The lack of mechanisms to consolidate and reuse error-specific experience causes repeated failures on similar challenging cases. To address these limitations, we propose SARE, a Sample-wise Adaptive textbfREasoning framework for training-free FGVR. Specifically, SARE adopts a cascaded design that combines fast candidate retrieval with fine-grained reasoning, invoking the latter only when necessary. In the reasoning process, SARE incorporates a self-reflective experience mechanism that leverages past failures to provide transferable discriminative guidance during inference, without any parameter updates. Extensive experiments across 14 datasets substantiate that SARE achieves state-of-the-art performance while substantially reducing computational overhead.

While Multimodal Large Language Models (MLLMs) have experienced rapid advancements, their visual encoders frequently remain a performance bottleneck. Conventional CLIP-based encoders struggle with dense spatial tasks due to the loss of visual details caused by low-resolution pretraining and the reliance on noisy, coarse web-crawled image-text pairs. To overcome these limitations, we introduce FineViT, a novel vision encoder specifically designed to unlock fine-grained perception. By replacing coarse web data with dense recaptions, we systematically mitigate information loss through a progressive training paradigm.: first, the encoder is trained from scratch at a high native resolution on billions of global recaptioned image-text pairs, establishing a robust, detail rich semantic foundation. Subsequently, we further enhance its local perception through LLM alignment, utilizing our curated FineCap-450M dataset that comprises over $450$ million high quality local captions. Extensive experiments validate the effectiveness of the progressive strategy. FineViT achieves state-of-the-art zero-shot recognition and retrieval performance, especially in long-context retrieval, and consistently outperforms multimodal visual encoders such as SigLIP2 and Qwen-ViT when integrated into MLLMs. We hope FineViT could serve as a powerful new baseline for fine-grained visual perception.

Affordance prediction serves as a critical bridge between perception and action in embodied AI. However, existing research is confined to pinhole camera models, which suffer from narrow Fields of View (FoV) and fragmented observations, often missing critical holistic environmental context. In this paper, we present the first exploration into Panoramic Affordance Prediction, utilizing 360-degree imagery to capture global spatial relationships and holistic scene understanding. To facilitate this novel task, we first introduce PAP-12K, a large-scale benchmark dataset containing over 1,000 ultra-high-resolution (12k, 11904 x 5952) panoramic images with over 12k carefully annotated QA pairs and affordance masks. Furthermore, we propose PAP, a training-free, coarse-to-fine pipeline inspired by the human foveal visual system to tackle the ultra-high resolution and severe distortion inherent in panoramic images. PAP employs recursive visual routing via grid prompting to progressively locate targets, applies an adaptive gaze mechanism to rectify local geometric distortions, and utilizes a cascaded grounding pipeline to extract precise instance-level masks. Experimental results on PAP-12K reveal that existing affordance prediction methods designed for standard perspective images suffer severe performance degradation and fail due to the unique challenges of panoramic vision. In contrast, PAP framework effectively overcomes these obstacles, significantly outperforming state-of-the-art baselines and highlighting the immense potential of panoramic perception for robust embodied intelligence.

We introduce the first unified framework for Fine-Grained Domain-Generalized Generalized Category Discovery (FG-DG-GCD), bringing open-world recognition closer to real-world deployment under domain shift. Unlike conventional GCD, which assumes labeled and unlabeled data come from the same distribution, DG-GCD learns only from labeled source data and must both recognize known classes and discover novel ones in unseen, unlabeled target domains. This problem is especially challenging in fine-grained settings, where subtle inter-class differences and large intra-class variation make domain generalization significantly harder. To support systematic evaluation, we establish the first FG-DG-GCD benchmarks by creating identity-preserving painting and sketch domains for CUB-200-2011, Stanford Cars, and FGVC-Aircraft using controlled diffusion-adapter stylization. On top of this ,we propose FoCUS, a single-stage framework that combines Domain-Consistent Parts Discovery (DCPD) for geometry-stable part reasoning with Uncertainty-Aware Feature Augmentation (UFA) for confidence-calibrated feature regularization through uncertainty-guided perturbations. Extensive experiments show that FoCUS outperforms strong GCD, FG-GCD, and DG-GCD baselines by 3.28%, 9.68%, and 2.07%, respectively, in clustering accuracy on the proposed benchmarks. It also remains competitive on coarse-grained DG-GCD tasks while achieving nearly 3x higher computational efficiency than the current state of the art. ^[Code and datasets will be released upon acceptance.]

When MLLMs fail at Science, Technology, Engineering, and Mathematics (STEM) visual reasoning, a fundamental question arises: is it due to perceptual deficiencies or reasoning limitations? Through systematic scaling analysis that independently scales perception and reasoning components, we uncover a critical insight: scaling perception consistently outperforms scaling reasoning. This reveals perception as the true lever limiting current STEM visual reasoning. Motivated by this insight, our work focuses on systematically enhancing the perception capabilities of MLLMs by establishing code as a powerful perceptual medium--executable code provides precise semantics that naturally align with the structured nature of STEM visuals. Specifically, we construct ICC-1M, a large-scale dataset comprising 1M Image-Caption-Code triplets that materializes this code-as-perception paradigm through two complementary approaches: (1) Code-Grounded Caption Generation treats executable code as ground truth for image captions, eliminating the hallucinations inherent in existing knowledge distillation methods; (2) STEM Image-to-Code Translation prompts models to generate reconstruction code, mitigating the ambiguity of natural language for perception enhancement. To validate this paradigm, we further introduce STEM2Code-Eval, a novel benchmark that directly evaluates visual perception in STEM domains. Unlike existing work relying on problem-solving accuracy as a proxy that only measures problem-relevant understanding, our benchmark requires comprehensive visual comprehension through executable code generation for image reconstruction, providing deterministic and verifiable assessment. Code is available at https://github.com/TongkunGuan/Qwen-CodePercept.

Multimodal large language models (MLLMs) have achieved remarkable performance across a wide range of vision language tasks. However, their ability in low-level visual perception, particularly in detecting fine-grained visual discrepancies, remains underexplored and lacks systematic analysis. In this work, we introduce OddGridBench, a controllable benchmark for evaluating the visual discrepancy sensitivity of MLLMs. OddGridBench comprises over 1,400 grid-based images, where a single element differs from all others by one or multiple visual attributes such as color, size, rotation, or position. Experiments reveal that all evaluated MLLMs, including open-source families such as Qwen3-VL and InternVL3.5, and proprietary systems like Gemini-2.5-Pro and GPT-5, perform far below human levels in visual discrepancy detection. We further propose OddGrid-GRPO, a reinforcement learning framework that integrates curriculum learning and distance-aware reward. By progressively controlling the difficulty of training samples and incorporating spatial proximity constraints into the reward design, OddGrid-GRPO significantly enhances the model's fine-grained visual discrimination ability. We hope OddGridBench and OddGrid-GRPO will lay the groundwork for advancing perceptual grounding and visual discrepancy sensitivity in multimodal intelligence. Code and dataset are available at https://wwwtttjjj.github.io/OddGridBench/.

Recent advances in multimodal large language models largely rely on CLIP-based visual encoders, which emphasize global semantic alignment but struggle with fine-grained visual understanding. In contrast, DINOv3 provides strong pixel-level perception yet lacks coarse-grained semantic abstraction, leading to limited multi-granularity reasoning. To address this gap, we propose Granulon, a novel DINOv3-based MLLM with adaptive granularity augmentation. Granulon introduces a text-conditioned granularity Controller that dynamically adjusts the visual abstraction level according to the semantic scope of the textual input, and an Adaptive Token Aggregation module that performs granularity-guided pooling and relation-aware clustering to produce compact, semantically rich visual tokens. This design enables unified "pixel-to-fine-to-coarse" reasoning within a single forward pass. Extensive and interpretable experiments demonstrate that Granulon improves accuracy by ~30% and reduces hallucination by ~20%, outperforming all visual encoders under identical settings.

The ability to distinguish subtle differences between visually similar images is essential for diverse domains such as industrial anomaly detection, medical imaging, and aerial surveillance. While comparative reasoning benchmarks for vision-language models (VLMs) have recently emerged, they primarily focus on images with large, salient differences and fail to capture the nuanced reasoning required for real-world applications. In this work, we introduce VLM-SubtleBench, a benchmark designed to evaluate VLMs on subtle comparative reasoning. Our benchmark covers ten difference types - Attribute, State, Emotion, Temporal, Spatial, Existence, Quantity, Quality, Viewpoint, and Action - and curate paired question-image sets reflecting these fine-grained variations. Unlike prior benchmarks restricted to natural image datasets, our benchmark spans diverse domains, including industrial, aerial, and medical imagery. Through extensive evaluation of both proprietary and open-source VLMs, we reveal systematic gaps between model and human performance across difference types and domains, and provide controlled analyses highlighting where VLMs' reasoning sharply deteriorates. Together, our benchmark and findings establish a foundation for advancing VLMs toward human-level comparative reasoning.

Fine-grained visual understanding is shifting from static classification to knowledge-augmented reasoning, where models must justify as well as recognise. Existing approaches remain limited by closed-set taxonomies and single-label prediction, leading to significant degradation under open-set or context-dependent conditions. We present the Knowledge-Augmented Fine-Grained Reasoning Agent (KFRA), a unified framework that transforms fine-grained perception into evidence-driven reasoning. KFRA operates through a three-stage closed reasoning loop that emulates expert analysis. It first performs open-vocabulary detection and web-scale retrieval to generate category hypotheses. It then conducts discriminative regions localisation by aligning textual knowledge with visual evidence through a global-to-local focusing mechanism. Finally, it integrates all multimodal evidence within a large multimodal model to perform interpretable reasoning. Unlike existing agents that treat retrieval and reasoning as independent processes, KFRA establishes a retrieval-grounding coupling that converts retrieved knowledge into spatially grounded evidence for verification. This design enables factual, interpretable, and task-agnostic reasoning across diverse fine-grained scenarios. To evaluate this capability, we construct FGExpertBench, a benchmark designed to assess reasoning depth and cross-task generalisation across six knowledge dimensions. Extensive experiments demonstrate that KFRA consistently surpasses both standalone large multimodal models and current agent frameworks, achieving up to 19 percent improvement in reasoning accuracy and delivering evidence-grounded interpretability in open-set fine-grained visual understanding.

A high-performing, general-purpose visual understanding model should map visual inputs to a taxonomic tree of labels, identify novel categories beyond the training set for which few or no publicly available images exist. Large Multimodal Models (LMMs) have achieved remarkable progress in fine-grained visual recognition (FGVR) for known categories. However, they remain limited in hierarchical visual recognition (HVR) that aims at predicting consistent label paths from coarse to fine categories, especially for novel categories. To tackle these challenges, we propose Taxonomy-Aware Representation Alignment (TARA), a simple yet effective strategy to inject taxonomic knowledge into LMMs. TARA leverages representations from biology foundation models (BFMs) that encode rich biological relationships through hierarchical contrastive learning. By aligning the intermediate representations of visual features with those of BFMs, LMMs are encouraged to extract discriminative visual cues well structured in the taxonomy tree. Additionally, we align the representations of the first answer token with the ground-truth label, flexibly bridging the gap between contextualized visual features and categories of varying granularity according to user intent. Experiments demonstrate that TARA consistently enhances LMMs' hierarchical consistency and leaf node accuracy, enabling reliable recognition of both known and novel categories within complex biological taxonomies. Code is available at https://github.com/PKU-ICST-MIPL/TARA_CVPR2026.

Complex video reasoning, actually, relies excessively on fine-grained perception rather than on expert (e.g., Ph.D, Science)-level reasoning. Through extensive empirical observation, we have recognized the critical impact of perception. In particular, when perception ability is almost fixed, enhancing reasoning from Qwen3-8B to OpenAI-o3 yields only 0.7% performance improvement. Conversely, even minimal change in perception model scale (from 7B to 32B) boosts performance by 1.4%, indicating enhancing perception, rather than reasoning, is more critical to improve performance. Therefore, exploring how to enhance perception ability through reasoning without the need for expensive fine-grained annotation information is worthwhile. To achieve this goal, we specially propose APPO, the Attention-guided Perception Policy Optimization algorithm that leverages token-level dense rewards to improve model's fine-grained perception. The core idea behind APPO is to optimize those tokens from different responses that primarily focus on the same crucial video frame (called intra-group perception tokens). Experimental results on diverse video benchmarks and models with different scales (3/7B) demonstrate APPO consistently outperforms GRPO and DAPO (0.5%~4%). We hope our work provides a promising approach to effectively enhance model's perception abilities through reasoning in a low-cost manner, serving diverse scenarios and demands.

Multimodal Large Language Models (MLLMs) are shifting towards "Thinking with Images" by actively exploring image details. While effective, large-scale training is computationally expensive, which has spurred growing interest in lightweight, training-free solutions. However, existing training-free methods suffer from two flaws: perceptual redundancy from indiscriminate cropping, which increases computational cost and introduces noise; and a drift between semantic intent and spatial attention, which prevents accurate localization of user-focused regions. To address these challenges, we propose LookWise, a framework for adaptive visual reasoning. LookWise follows a two-stage pipeline: a confidence-based module decides when to look more carefully, and a semantic-guided localization module determines where to look. This design enables MLLMs to adaptively acquire fine-grained visual evidence without additional training. Experiments on fine-grained and high-resolution visual reasoning benchmarks show that LookWise consistently improves accuracy over strong baselines while achieving an approximately $4.0\times$ inference speedup over the search-based method ZoomEye, demonstrating robust cross-model generalization.

Following the success of Group Relative Policy Optimization (GRPO) in foundation LLMs, an increasing number of works have sought to adapt GRPO to Visual Large Language Models (VLLMs) for visual perception tasks (e.g., detection and segmentation). However, much of this line of research rests on a long-standing yet unexamined assumption: training paradigms developed for language reasoning can be transferred seamlessly to visual perception. Our experiments show that this assumption is not valid, revealing intrinsic differences between reasoning-oriented and perception-oriented settings. Using reasoning segmentation as a representative case, we surface two overlooked factors: (i) the need for a broader output space, and (ii) the importance of fine-grained, stable rewards. Building on these observations, we propose Dr.~Seg, a simple, plug-and-play GRPO-based framework consisting of a Look-to-Confirm mechanism and a Distribution-Ranked Reward module, requiring no architectural modifications and integrating seamlessly with existing GRPO-based VLLMs. Extensive experiments demonstrate that Dr.~Seg improves performance in complex visual scenarios while maintaining strong generalization. Code, models, and datasets are available at https://github.com/eVI-group-SCU/Dr-Seg.

Vision Transformers, ViTs, have emerged as a powerful alternative to convolutional neural networks, CNNs, in a variety of image-based tasks. While CNNs have previously been evaluated for their ability to perform graphical perception tasks, which are essential for interpreting visualizations, the perceptual capabilities of ViTs remain largely unexplored. In this work, we investigate the performance of ViTs in elementary visual judgment tasks inspired by the foundational studies of Cleveland and McGill, which quantified the accuracy of human perception across different visual encodings. Inspired by their study, we benchmark ViTs against CNNs and human participants in a series of controlled graphical perception tasks. Our results reveal that, although ViTs demonstrate strong performance in general vision tasks, their alignment with human-like graphical perception in the visualization domain is limited. This study highlights key perceptual gaps and points to important considerations for the application of ViTs in visualization systems and graphical perceptual modeling.

Vision-language models (VLMs) have made substantial progress across a wide range of visual question answering benchmarks, spanning visual reasoning, document understanding, and multimodal dialogue. These improvements are evident in a wide range of VLMs built on a variety of base models, alignment architectures, and training data. However, recent works show that these models trail behind in traditional image classification benchmarks, which test fine-grained visual knowledge. We test a large number of recent VLMs on fine-grained classification benchmarks and identify potential factors in the disconnect between fine-grained knowledge and other vision benchmarks. Through a series of ablation experiments, we find that using a better LLM improves all benchmark scores equally, while a better vision encoder disproportionately improves fine-grained classification performance. Furthermore, we find that the pretraining stage is also vital to fine-grained performance, particularly when the language model weights are unfrozen during pretraining. These insights pave the way for enhancing fine-grained visual understanding and vision-centric capabilities in VLMs.

Modern vision models increasingly rely on rich semantic representations that extend beyond class labels to include descriptive concepts, attributes, and contextual cues. However, semantic concepts are not uniformly represented across classes: a concept may be frequent globally yet remain underrepresented within a class, resulting in low class-concept coverage. We formalize this class-concept phenomenon as Semantic Coverage Imbalance (SCI) and study its relationship with learned semantic representations. We find that lower class-concept coverage is consistently associated with weaker semantic representations. To address this, we introduce Coverage-aware Semantic Representation Learning, which shares more semantic structure for low coverage or uncertain relations while preserving more pair-specific structure for high coverage relations. We instantiate this principle in SemCovNet with Coverage-Calibrated Semantic Sharing (CCSS), a coverage- and uncertainty-aware partial-pooling mechanism. Using a range of vision datasets spanning facial attributes and fine-grained recognition, as well as real-world medical datasets, we show that SCI is widespread, degrades representation quality, and can be mitigated through coverage-aware semantic representation learning.

While Large Vision-Language Models (LVLMs) have demonstrated remarkable capabilities for reasoning and self-correction at the textual level, these strengths provide minimal benefits for complex tasks centered on visual perception, such as Chart Parsing. Existing models often struggle with visually dense charts, leading to errors like data omission, misalignment, and hallucination. Inspired by the human strategy of using a finger as a ``visual anchor'' to ensure accuracy when reading complex charts, we propose a new paradigm named Visual Self-Refine (VSR). The core idea of VSR is to enable a model to generate pixel-level localization outputs, visualize them, and then feed these visualizations back to itself, allowing it to intuitively inspect and correct its own potential visual perception errors. We instantiate the VSR paradigm in the domain of Chart Parsing by proposing ChartVSR. This model decomposes the parsing process into two stages: a Refine Stage, where it iteratively uses visual feedback to ensure the accuracy of all data points' Pixel-level Localizations, and a Decode Stage, where it uses these verified localizations as precise visual anchors to parse the final structured data. To address the limitations of existing benchmarks, we also construct ChartP-Bench, a new and highly challenging benchmark for chart parsing. Our work also highlights VSR as a general-purpose visual feedback mechanism, offering a promising new direction for enhancing accuracy on a wide range of vision-centric tasks.

Fine-grained visual reasoning in multimodal large language models (MLLMs) is bottlenecked by single-pass global image encoding: key evidence often lies in tiny objects, cluttered regions, subtle markings, or dense charts. We present TikArt (Thinking Aperture), an aperture-guided agent that formulates multimodal reasoning as sequential evidence acquisition over regions of interest. TikArt follows a Think--Aperture--Observe (TAO) loop that interleaves language reasoning with two aperture actions: Zoom, which extracts rectangular crops, and Segment, which invokes an off-the-shelf segmenter to produce object-centric mask-based views for irregular targets. A mandatory Observation step after every aperture action writes local evidence back into text, yielding interpretable aperture trajectories and persistent linguistic memory. Built on Qwen3-VL-8B, TikArt is trained with GRPO-style reinforcement learning under a two-stage curriculum. To stabilize long-horizon tool-integrated learning, we introduce Relative Uncertainty Reduction (RUR), a dense reward computed by a frozen evaluator that favors evidence-building trajectories and mitigates degenerate tool use. Experiments on high-resolution reasoning, general multimodal understanding, and both referring and reasoning-oriented segmentation show consistent gains over the backbone, demonstrating that aperture-guided observation improves fine-grained visual reasoning and transfers naturally to pixel-level grounding.