影响力指数

1.39/100

超过 3.4%

全站排名 #62,198

发表论文3 篇

平均评分4.0

年均产出3.0 篇/年

Ma Shuguang

Researcher@Alibaba Group·中国·OpenReview

研究方向

large language model

MAPO: Momentum-Aware Policy Optimization

ICLR 2026Withdrawn

Selective Expert Guidance for Effective and Diverse Exploration in Reinforcement Learning of LLMs

ICLR 2026Poster

HINT: Helping Ineffective Rollouts Navigate Towards Effectiveness

ICLR 2026Withdrawn

合作者 (17)