Saved in:
Bibliographic Details
Main Authors: Meng, Fanqing, Du, Lingxiao, Gu, Jiawei, Liao, Jiaqi, Li, Linjie, Wu, Zijian, Liu, Xiangyan, Zhao, Ziqi, Hu, Mengkang, Liu, Zichen, Zhang, Jiaheng, Shieh, Michael Qizhe
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.15432
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915923204505600
author Meng, Fanqing
Du, Lingxiao
Gu, Jiawei
Liao, Jiaqi
Li, Linjie
Wu, Zijian
Liu, Xiangyan
Zhao, Ziqi
Hu, Mengkang
Liu, Zichen
Zhang, Jiaheng
Shieh, Michael Qizhe
author_facet Meng, Fanqing
Du, Lingxiao
Gu, Jiawei
Liao, Jiaqi
Li, Linjie
Wu, Zijian
Liu, Xiangyan
Zhao, Ziqi
Hu, Mengkang
Liu, Zichen
Zhang, Jiaheng
Shieh, Michael Qizhe
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2603_15432
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Gym-V: A Unified Vision Environment System for Agentic Vision Research
Meng, Fanqing
Du, Lingxiao
Gu, Jiawei
Liao, Jiaqi
Li, Linjie
Wu, Zijian
Liu, Xiangyan
Zhao, Ziqi
Hu, Mengkang
Liu, Zichen
Zhang, Jiaheng
Shieh, Michael Qizhe
Computer Vision and Pattern Recognition
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.
title Gym-V: A Unified Vision Environment System for Agentic Vision Research
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2603.15432