G1: Bootstrapping Perception and Reasoning Abilities of Vision-Language Model via Reinforcement Learning
This addresses the problem of enabling VLMs to act as autonomous agents in interactive environments, representing a novel method for a known bottleneck rather than an incremental improvement.
The paper tackles the 'knowing-doing' gap in Vision-Language Models (VLMs), where they perform poorly in interactive visual environments like games, by introducing VLM-Gym for RL training and G1 models that outperform leading proprietary models such as Claude-3.7-Sonnet-Thinking across all games.
Vision-Language Models (VLMs) excel in many direct multimodal tasks but struggle to translate this prowess into effective decision-making within interactive, visually rich environments like games. This ``knowing-doing'' gap significantly limits their potential as autonomous agents, as leading VLMs often performing badly in simple games. To address this, we introduce VLM-Gym, a curated reinforcement learning (RL) environment featuring diverse visual games with unified interfaces and adjustable, compositional difficulty, specifically designed for scalable multi-game parallel training. Leveraging VLM-Gym, we train G0 models using pure RL-driven self-evolution, which demonstrate emergent perception and reasoning patterns. To further mitigate challenges arising from game diversity, we develop G1 models. G1 incorporates a perception-enhanced cold start prior to RL fine-tuning. Our resulting G1 models consistently surpass their teacher across all games and outperform leading proprietary models like Claude-3.7-Sonnet-Thinking. Systematic analysis reveals an intriguing finding: perception and reasoning abilities mutually bootstrap each other throughout the RL training process. Source code including VLM-Gym and RL training are released at https://github.com/chenllliang/G1 to foster future research in advancing VLMs as capable interactive agents.