paper-with-me

홈 › Papers

Gym-V: A Unified Vision Environment System for Agentic Vision Research

2026-03-16 · Fanqing Meng, Lingxiao Du, Jiawei Gu, Jiaqi Liao, Linjie Li, Zijian Wu, Xiangyan Liu, Ziqi Zhao, Mengkang Hu, Zichen Liu, Jiaheng Zhang, Michael Qizhe Shieh arxiv

As agentic systems increasingly rely on reinforcement learning from verifiable rewards, standardized ``gym'' infrastructure has become essential for rapid iteration, reproducibility, and fair comparison. Vision agents lack such infrastructure, limiting systematic study of what drives their learning and where current models fall short. We introduce \textbf{Gym-V}, a unified platform of 179 procedurally generated visual environments across 10 domains with controllable difficulty, enabling controlled experiments that were previously infeasible across fragmented toolkits. Using it, we find that observation scaffolding is more decisive for training success than the choice of RL algorithm, with captions and game rules determining whether learning succeeds at all. Cross-domain transfer experiments further show that training on diverse task categories generalizes broadly while narrow training can cause negative transfer, with multi-turn interaction amplifying all of these effects. Gym-V is released as a convenient foundation for training environments and evaluation toolkits, aiming to accelerate future research on agentic VLMs.

📄 PDF Abstract BibTeX arXiv:2603.15432

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement Learning

Similar Papers 제목 키워드 기반

PyVision-RL: Forging Open Agentic Vision Models via RL

2026-02-24 · Shitian Zhao, Shaoheng Lin, Ming Li, Haoquan Zhang 외 arxiv

Reinforcement learning for agentic multimodal models often suffers from interaction collapse, where models learn to reduce tool usage and multi-turn reasoning, limiting the benefits of agentic behavior. We introduce PyVi…

Reinforcement Learning

Hybrid Advantage Estimation with Unified Critic for VLM Agentic Reinforcement Learning

2026-07-26 · Wenxuan Zhang, Yuhui Wang, Donggang Jia, Xiaoqian Shen 외 arxiv

Large Vision-Language Models (VLMs) now act as agents in interactive environments, where success requires coherent reasoning and decision-making across turns. Although end-to-end training in agentic environments can impr…

Reinforcement Learning

Autonomous Computer Vision Development with Agentic AI

2025-06-11 · Jin Kim, Muhammad Wahi-Anwa, Sangyun Park, Shawn Shin 외

Agentic Artificial Intelligence (AI) systems leveraging Large Language Models (LLMs) exhibit significant potential for complex reasoning, planning, and tool utilization. We demonstrate that a specialized computer vision …

Medical Image Analysis

AgenticDiffusion: Agentic Diffusion-based Path Planning for Vision-Based UAV Navigation

2026-06-02 · Faryal Batool, Muhammad Ahsan Mustafa, Fawad Mehboob, Valerii Serpiva 외 arxiv

Indoor UAV navigation requires efficient exploration, scene understanding, and reliable trajectory execution under limited field-of-view observations. Existing vision-based navigation frameworks typically rely on single-…

Scene Understanding

Agentic Reasoning and Tool Integration for LLMs via Reinforcement Learning

2025-04-28 · Joykirat Singh, Raghav Magazine, Yash Pandya, Akshay Nambi

Large language models (LLMs) have achieved remarkable progress in complex reasoning tasks, yet they remain fundamentally limited by their reliance on static internal knowledge and text-only reasoning. Real-world problem …

Mathematical Reasoning