影响力指数
论文质量、代表作、近期表现、广度与样本量置信度综合计算
43.76/100
前 11.8%
全站排名 #7,582
发表论文8 篇
平均评分
年均产出2.7 篇/年
Bei Li
研究方向
DPO/PPO algorithms · Large Language Models · Architecture Improvement · Machine Translation · Sequence Generation · Deep Transformer models
39
MRO: Enhancing Reasoning in Diffusion Language Models via Multi-Reward Optimization
NeurIPS 2025Poster
30
Earlier Tokens Contribute More: Learning Direct Preference Optimization From Temporal Decay Perspective
ICLR 2025Poster
二作11
GRAM: A Generative Foundation Reward Model for Reward Generalization
ICML 2025Poster
14
InteractiveCOT: Aligning Dynamic Chain-of-Thought Planning for Embodied Decision-Making
ICLR 2025Rejected