Benchmarking Large Vision-Language Models on Fine-Grained Image Tasks: A Comprehensive Evaluation

[2504.14988] Benchmarking Large Vision-Language Models on Fine-Grained Image Tasks: A Comprehensive Evaluation Skip to main content Search Submit Donate Log in Search arXiv Press Enter to search · Advanced search Computer Science > Computer Vision and Pattern Recognition arXiv:2504.14988 (cs) [Submitted on 21 Apr 2025 ( v1 ), last revised 30 Apr 2026 (this version, v4)] Title: Benchmarking Large Vision-Language Models on Fine-Grained Image Tasks: A Comprehensive Evaluation Authors: Hong-Tao Yu , Yuxin Peng , Serge Belongie , Xiu-Shen Wei View a PDF of the paper titled Benchmarking Large Vision-Language Models on Fine-Grained Image Tasks: A Comprehensive Evaluation, by Hong-Tao Yu and Yuxin Peng and Serge Belongie and Xiu-Shen Wei View PDF HTML (experimental) Abstract: Recent advancements in Large Vision-Language Models (LVLMs) have demonstrated remarkable multimodal perception capabilities, garnering significant attention. While numerous evaluation studies have emerged, assessing LVLMs both holistically and on specialized tasks, fine-grained image tasks-fundamental to computer vision-remain largely unexplored. To fill this gap, we introduce a comprehensive fine-grained evaluation benchmark, i.e., FG-BMK, comprising 1.01 million questions and 0.33 million images. Our evaluation systematically examines LVLMs from both human-oriented and machine-oriented perspectives, focusing on their semantic recognition and fine-grained feature representation capabilities. Through extensive experiments on twelve representative LVLMs/VLMs, we uncover key findings regarding the influence of training paradigms, modality alignment, perturbation susceptibility, and fine-grained category reasoning on task performance. This work provides critical insights into the limitations of current LVLMs and offers guidance for future data construction and model design in the development of more advanced LVLMs. Our code is open-source and available at this https URL . Comments: Accepted to ICLR 2026 Subjects: Computer Vision and Pattern Recognition (cs.CV) Cite as: arXiv:2504.14988 [cs.CV] (or arXiv:2504.14988v4 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2504.14988 Focus to learn more arXiv-issued DOI via DataCite Submission history From: Hongtao Yu [ view email ] [v1] Mon, 21 Apr 2025 09:30:41 UTC (4,752 KB) [v2] Wed, 14 May 2025 02:33:48 UTC (6,249 KB) [v3] Sat, 11 Apr 2026 13:42:15 UTC (2,238 KB) [v4] Thu, 30 Apr 2026 15:12:07 UTC (2,238 KB) Full-text links: Access Paper: View a PDF of the paper titled Benchmarking Large Vision-Language Models on Fine-Grained Image Tasks: A Comprehensive Evaluation, by Hong-Tao Yu and Yuxin Peng and Serge Belongie and Xiu-Shen Wei View PDF HTML (experimental) TeX Source view license Current browse context: cs.CV < prev | next > new | recent | 2025-04 Change to browse by: cs References & Citations NASA ADS Google Scholar Semantic Scholar export BibTeX citation Loading... BibTeX formatted citation × loading... Data provided by: Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer ( What is the Explorer? ) Connected Papers Toggle Connected Papers ( What is Connected Papers? ) Litmaps Toggle Litmaps ( What is Litmaps? ) scite.ai Toggle scite Smart Citations ( What are Smart Citations? ) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv ( What is alphaXiv? ) Links to Code Toggle CatalyzeX Code Finder for Papers ( What is CatalyzeX? ) DagsHub Toggle DagsHub ( What is DagsHub? ) GotitPub Toggle Gotit.pub ( What is GotitPub? ) Huggingface Toggle Hugging Face ( What is Huggingface? ) ScienceCast Toggle ScienceCast ( What is ScienceCast? ) Demos Demos Replicate Toggle Replicate ( What is Replicate? ) Spaces Toggle Hugging Face Spaces ( What is Spaces? ) Spaces Toggle TXYZ.AI ( What is TXYZ.AI? ) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower ( What are Influence Flowers? ) Core recommender toggle CORE Recommender ( What is CORE? ) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs . Which authors of this paper are endorsers? | Disable MathJax ( What is MathJax? ) We gratefully acknowledge support from our major funders , member institutions , , and all contributors. About · Help · Contact · Subscribe · Copyright · Privacy · Accessibility · Operational Status (opens in new tab) Major funding support from

Related documents

• A Unified Hierarchical Framework for Fine-grained Cross-view Geo-localization over Large-scale Scenarios#1012017
• From Glance to Scrutiny: Progressive Distortion Reasoning for Fine-Grained Image Quality Assessment#985680
• VISTA-Bench: Benchmarking Multilingual Image Translation with Image-Specific Rubrics#1141330
• CGC: Compositional Grounded Contrast for Fine-Grained Multi-Image Understanding#134586
• PathView-Bench: Can Multimodal Large Language Models Achieve Fine-grained Multiscale Understanding of Pathology Images?#414159