paper-with-me

홈 › Papers

B-GRTO: Bootstrapped Group Relative Tool Optimization for Referring Segmentation

2026-05-22 · Mario Markov, Stefan Maria Ailuro, Mohammad Mahdi, Luc Van Gool, Danda Pani Paudel arxiv

Segmentation is a fundamental task in computer vision, underpinning pixel-level scene understanding and serving as a cornerstone for applications ranging from autonomous perception to medical image analysis. For complex referring segmentation, recent methods pair large vision-language models with segmentation decoders: the former analyzes the image and prompt, while the latter predicts the target mask. Although reinforcement learning improves reasoning-intensive vision-language systems, trainable tools such as segmentation decoders are typically optimized separately with differentiable objectives, and the principled integration of such objectives into reinforcement learning remains underexplored. Thus, we introduce group relative tool optimization (GRTO), a mathematically grounded framework for jointly optimizing a policy with differentiable tool use. GRTO reuses group relative policy optimization (GRPO) rollouts to optimize the auxiliary tool objective, letting decoder gradients complement policy rewards. Further, we derive Bootstrapped-GRTO (B-GRTO), a pre-training method that cheaply bootstraps the tool, leading to faster convergence and superior performance. Across three challenging referring segmentation settings, B-GRTO results in substantial improvements over plain GRPO, matching or surpassing domain-specific state-of-the-art methods. This demonstrates the value of unifying reinforcement learning with differentiable auxiliary objectives for reasoning-intensive segmentation.

📄 PDF Abstract BibTeX arXiv:2605.23500

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement LearningScene Understanding

Similar Papers 제목 키워드 기반

Testing separability and independence of perceptual dimensions with general recognition theory: A tutorial and new R package (grtools)

2016-10-11

Determining whether perceptual properties are processed independently is an important goal in perceptual science, and tools to test independence should be widely available to experimental researchers. The best analytical…

Experimental Design

Bootstrapped Mixed Rewards for RL Post-Training: Injecting Canonical Action Order

2025-12-03 · Prakhar Gupta, Vaibhav Gupta arxiv

Post-training with reinforcement learning (RL) typically optimizes a single scalar objective and ignores structure in how solutions are produced. We ask whether a scalar hint toward a canonical solver ordering, used only…

Reinforcement Learning

Hybrid Group Relative Policy Optimization: A Multi-Sample Approach to Enhancing Policy Optimization

2025-01-30 · Soham Sane

Hybrid Group Relative Policy Optimization (Hybrid GRPO) is a reinforcement learning framework that extends Proximal Policy Optimization (PPO) and Group Relative Policy Optimization (GRPO) by incorporating empirical multi…

reinforcement-learningReinforcement Learning

Advancing SLM Tool-Use Capability using Reinforcement Learning

2025-09-03 · Dhruvi Paprunia, Vansh Kharidia, Pankti Doshi arxiv

In an era where tool-augmented AI agents are becoming increasingly vital, our findings highlight the ability of Group Relative Policy Optimization (GRPO) to empower SLMs, which are traditionally constrained in tool use. …

Reinforcement Learning

RC-GRPO: Reward-Conditioned Group Relative Policy Optimization for Multi-Turn Tool Calling Agents

2026-02-03 · Haitian Zhong, Jixiu Zhai, Lei Song, Jiang Bian 외 arxiv

Multi-turn tool calling is challenging for Large Language Models (LLMs) because rewards are sparse and exploration is expensive. A common recipe, SFT followed by GRPO, can stall when within-group reward variation is low …