← Search

Yinsong Ma

2 accepted papers

2026

Play to Generalize: Learning to Reason Through Game Play

ICLR 2026poster

Developing reasoning capabilities in multimodal large language models (MLLMs) remains challenging. Motivated by literature suggesting that gameplay promotes transferable reasoning skills, we propose a novel post-training method, Visual Game Learning (ViGaL), where MLLMs develop generalizable reasoni…

Cited by 0SourcecodeScholar
2025

LipsNet++: Unifying Filter and Controller into a Policy Network

ICML 2025spotlight

Deep reinforcement learning (RL) is effective for decision-making and control tasks like autonomous driving and embodied AI. However, RL policies often suffer from the action fluctuation problem in real-world applications, resulting in severe actuator wear, safety risk, and performance degradation.…