影响力指数

-/100

发表论文2 篇

平均评分4.5

年均产出2.0 篇/年

Jiabin Fan

Researcher@Tencent·中国·OpenReview

研究方向

Reinforcement Learning · Machine Learning · Learning to Rank

BRIM: Block-wise Return Induction Method for Sequence Knowledge Distillation

ICLR 2026Withdrawn

Reward Model Boosting for RLHF

ICLR 2026Rejected

合作者 (5)

Michael Bowling