Reinforcing VLMs to Use Tools for Detailed Visual Reasoning Under Resource Constraints

• URL: http://arxiv.org/abs/2506.14821v3
• Date: Tue, 05 Aug 2025 03:49:33 GMT
• Title: Reinforcing VLMs to Use Tools for Detailed Visual Reasoning Under Resource Constraints
• Authors: Sunil Kumar, Bowen Zhao, Leo Dirac, Paulina Varshavskaya,
• Abstract summary: We draw inspiration from methods like Deepseek-r1 for vision-language models (VLMs) and train smaller-scale models with Group Relative Policy Optimization (GRPO) to use external tools such as zoom.<n>The greatest benefit is obtained with a combination of GRPO learning, a simple reward structure, a simplified tool-calling interface, and a training data mix that over-represents visually difficult examples.<n>Compared to similarly-sized baseline models, our method achieves better performance on some visual question-answering (VQA) tasks, thanks to the detailed visual information gathered from the external tool.
• Score: 8.411630512737887
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: Despite tremendous recent advances in large model reasoning ability, vision-language models (VLMs) still struggle with detailed visual reasoning, especially when compute resources are limited. To address this challenge, we draw inspiration from methods like Deepseek-r1 for VLMs and train smaller-scale models with Group Relative Policy Optimization (GRPO) to use external tools such as zoom. The greatest benefit is obtained with a combination of GRPO learning, a simple reward structure, a simplified tool-calling interface, allocating additional tokens to the result of the tool call, and a training data mix that over-represents visually difficult examples. Compared to similarly-sized baseline models, our method achieves better performance on some visual question-answering (VQA) tasks, thanks to the detailed visual information gathered from the external tool.

Related papers

• VRAG-RL: Empower Vision-Perception-Based RAG for Visually Rich Information Understanding via Iterative Reasoning with Reinforcement Learning [45.39372905700317]
  We introduce VRAG-RL, a novel RL framework tailored for complex reasoning across visually rich information.<n>With this framework, VLMs interact with search engines, autonomously sampling single-turn or multi-turn reasoning trajectories.<n>Our approach highlights key limitations of RL in RAG domains.
  arXiv  Detail & Related papers  (2025-05-28T06:30:51Z)
• VisualToolAgent (VisTA): A Reinforcement Learning Framework for Visual Tool Selection [47.259066449806866]
  VisTA is a new reinforcement learning framework that empowers visual agents to dynamically explore, select, and combine tools from a diverse library based on empirical performance.<n>We show that VisTA achieves substantial performance gains over training-free baselines.<n>These results highlight VisTA's ability to enhance generalization, adaptively utilize diverse tools, and pave the way for flexible, experience-driven visual reasoning systems.
  arXiv  Detail & Related papers  (2025-05-26T17:59:17Z)
• OpenThinkIMG: Learning to Think with Images via Visual Tool Reinforcement Learning [57.89304342666846]
  We introduce OpenThinkIMG, the first open-source, comprehensive end-to-end framework for tool-augmented LVLMs.<n>We propose a novel reinforcement learning framework V-ToolRL to train LVLMs to learn adaptive policies for invoking external vision tools.<n>V-ToolRL enables LVLMs to autonomously discover optimal tool-usage strategies.
  arXiv  Detail & Related papers  (2025-05-13T14:35:51Z)
• ToolRL: Reward is All Tool Learning Needs [54.16305891389931]
  Large Language Models (LLMs) often undergo supervised fine-tuning (SFT) to acquire tool use capabilities.<n>Recent advancements in reinforcement learning (RL) have demonstrated promising reasoning and generalization abilities.<n>We present the first comprehensive study on reward design for tool selection and application tasks within the RL paradigm.
  arXiv  Detail & Related papers  (2025-04-16T21:45:32Z)
• FamilyTool: A Multi-hop Personalized Tool Use Benchmark [93.80355496575281]
  FamilyTool is a benchmark grounded in a family-based knowledge graph (KG) that simulates personalized, multi-hop tool use scenarios.<n> Experiments reveal significant performance gaps in state-of-the-art Large Language Models (LLMs)<n>FamilyTool serves as a critical resource for evaluating and advancing LLM agents' reasoning, adaptability, and scalability in complex, dynamic environments.
  arXiv  Detail & Related papers  (2025-04-09T10:42:36Z)
• FOLDER: Accelerating Multi-modal Large Language Models with Enhanced Performance [9.782362715017596]
  We introduce FOLDER, a simple yet effective plug-and-play module designed to reduce the length of the visual token sequence.<n>We analyze the information loss introduced by different reduction strategies and develop FOLDER to preserve key information while removing visual redundancy.<n>FOLDER achieves comparable or even better performance than the original models, while dramatically reducing complexity by removing up to 70% of visual tokens.
  arXiv  Detail & Related papers  (2025-01-05T03:28:45Z)
• AvaTaR: Optimizing LLM Agents for Tool Usage via Contrastive Reasoning [93.96463520716759]
  Large language model (LLM) agents have demonstrated impressive capabilities in utilizing external tools and knowledge to boost accuracy and hallucinations. Here, we introduce AvaTaR, a novel and automated framework that optimize an LLM agent to effectively leverage provided tools, improving performance on a given task.
  arXiv  Detail & Related papers  (2024-06-17T04:20:02Z)
• VQA Training Sets are Self-play Environments for Generating Few-shot Pools [2.556825820539693]
  We propose a technique in which existing training sets can be directly used for constructing computational environments with task metrics as rewards. The proposed method starts with zero-shot prompts and iteratively refines them by selecting few-shot examples that maximize the task metric on the training set. Our experiments showcase how Gemini learns how to use itself, or another smaller and specialized model such as ScreenAI, to iteratively improve performance on training sets.
  arXiv  Detail & Related papers  (2024-05-30T07:38:58Z)
• Towards Completeness-Oriented Tool Retrieval for Large Language Models [60.733557487886635]
  Real-world systems often incorporate a wide array of tools, making it impractical to input all tools into Large Language Models. Existing tool retrieval methods primarily focus on semantic matching between user queries and tool descriptions. We propose a novel modelagnostic COllaborative Learning-based Tool Retrieval approach, COLT, which captures not only the semantic similarities between user queries and tool descriptions but also takes into account the collaborative information of tools.
  arXiv  Detail & Related papers  (2024-05-25T06:41:23Z)
• Chain-of-Spot: Interactive Reasoning Improves Large Vision-Language Models [81.71651422951074]
  Chain-of-Spot (CoS) method is a novel approach that enhances feature extraction by focusing on key regions of interest. This technique allows LVLMs to access more detailed visual information without altering the original image resolution. Our empirical findings demonstrate a significant improvement in LVLMs' ability to understand and reason about visual content.
  arXiv  Detail & Related papers  (2024-03-19T17:59:52Z)

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.