Towards High-Resolution Visual Perception via Hierarchical Entity Exploration

• URL: http://arxiv.org/abs/2607.00816v1
• Date: Wed, 01 Jul 2026 11:41:21 GMT
• Title: Towards High-Resolution Visual Perception via Hierarchical Entity Exploration
• Abstract summary: Hierarchical Entity Exploration transforms static image understanding into query-guided entity exploration.<n>HEE is a training-free and model-agnostic framework that transforms static image understanding into dynamic, query-guided entity exploration.
• Score: 36.649153640650496
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: High-resolution (HR) image perception remains a key challenge in multimodal large language models (MLLMs), as fine-grained details are often lost when the image is processed as a whole. Existing methods either require training to teach models where to look or heuristically divide the image into fixed regions, both of which struggle to generalize in complex HR scenes. In this work, we propose Hierarchical Entity Exploration (HEE), a training-free and model-agnostic framework that transforms static image understanding into dynamic, query-guided entity exploration. HEE first evaluates each region using a dual scoring mechanism to determine whether it already contains sufficient evidence to answer the question. If not, it applies object detection within the most promising region to extract fine-grained entities, clusters them into coherent subregions, and organizes them into a multi-level semantic hierarchy for deeper exploration. When deeper regions still fail to yield confident answers, a confidence-guided backtracking mechanism revisits alternative paths to ensure adaptive perception. Extensive results show that HEE outperforms training-free methods like ZoomEye and RAP in both accuracy and efficiency on two complex HR benchmarks (Visual Probe and HR-Bench), across different MLLMs such as Qwen2.5-VL and LLaVA-OneVision. Moreover, HEE demonstrates generalization on the MME-RealWorld benchmark.

Related papers

• EgoHieraLoc: A Cortically Inspired Hierarchical Segmentation-Guided Framework for Egocentric Visual Query Localization [9.547789251222838]
  Visual query localization (VQL) aims to retrieve and re-localize a queried object in egocentric videos.<n>Human vision handles ambiguity through a hierarchical process.<n>We propose textbfEgoHieraLoc, a unified framework for VQL-2D and VQL-3D.
  arXiv  Detail & Related papers  (2026-08-10T14:31:01Z)
• Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation [71.92541392470103]
  Multimodal Large Language Models (MLLMs) still struggle with fine-grained visual understanding.<n>We propose Vision-OPD (Vision On-Policy Distillation), a regional-to-global self-distillation framework.<n>Vision-OPD instantiates two conditional policies from the same MLLM: a crop-conditioned teacher and a full-image-conditioned student.
  arXiv  Detail & Related papers  (2026-05-18T17:57:04Z)
• Learning to Focus and Precise Cropping: A Reinforcement Learning Framework with Information Gaps and Grounding Loss for MLLMs [58.32326205291745]
  MLLMs autonomously utilize image cropping tool to analyze regions of interest for question answering.<n>We demonstrate the model's strong reliance on global input and its weak dependence on the details within the cropped region.<n>We propose a novel two-stage reinforcement learning framework that does not require supervision.
  arXiv  Detail & Related papers  (2026-03-29T03:18:57Z)
• Ego to World: Collaborative Spatial Reasoning in Embodied Systems via Reinforcement Learning [61.753025885751036]
  We introduce the Ego-to-World benchmark, which evaluates a vision-language model's ability to fuse heterogeneous viewpoints across three tasks.<n>We propose CoRL, a two-stage framework that combines Chain-of-Thought supervised fine-tuning with reinforcement learning.<n>We show that CoRL consistently surpasses strong proprietary and open-source baselines on both reasoning and perception-grounding metrics.
  arXiv  Detail & Related papers  (2026-03-16T04:27:53Z)
• TikArt: Aperture-Guided Observation for Fine-Grained Visual Reasoning via Reinforcement Learning [6.656456191281567]
  We introduce TikArt, an aperture-guided agent that casts vision-language reasoning as a decision process over regions of interest.<n>TikArt follows a Think-Aperture-Observe loop, alternating between language generation and two aperture actions.<n>After every action, the model must produce an explicit observation, turning local visual cues into persistent linguistic memory.
  arXiv  Detail & Related papers  (2026-02-16T05:46:47Z)
• DRFormer: A Dual-Regularized Bidirectional Transformer for Person Re-identification [37.61695934257133]
  Fine-grained discriminative details and global semantic features can contribute to solving person re-identification challenges.<n>Vision foundation models excel at mining local textures, and vision-language models capture strong global semantic difference.<n>We propose a framework to synergize their strengths by a textbfDual-textbfRegularized Bidirectional textbfTransformer.
  arXiv  Detail & Related papers  (2026-02-01T06:59:53Z)
• Revisiting Multi-Task Visual Representation Learning [52.93947931352643]
  We introduce MTV, a principled multi-task visual pretraining framework.<n>We leverage high-capacity "expert" models to synthesize dense, structured pseudo-labels at scale.<n>Our results demonstrate that MTV achieves "best-of-both-worlds" performance.
  arXiv  Detail & Related papers  (2026-01-20T11:59:19Z)
• FineRS: Fine-grained Reasoning and Segmentation of Small Objects with Reinforcement Learning [62.11389260206383]
  textscFineRS is a two-stage MLLM-based reinforcement learning framework for segmenting extremely small objects.<n>We present textscFineRS-4k, a new dataset for evaluating MLLMs on attribute-level reasoning and pixel-level segmentation on subtle, small-scale targets.
  arXiv  Detail & Related papers  (2025-10-24T10:14:17Z)
• ImageScope: Unifying Language-Guided Image Retrieval via Large Multimodal Model Collective Reasoning [62.61187785810336]
  ImageScope is a training-free, three-stage framework that unifies language-guided image retrieval tasks.<n>In the first stage, we improve the robustness of the framework by synthesizing search intents across varying levels of semantic granularity.<n>In the second and third stages, we reflect on retrieval results by verifying predicate propositions locally, and performing pairwise evaluations globally.
  arXiv  Detail & Related papers  (2025-03-13T08:43:24Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.