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

Kan Li

Full Professor@Beijing Institute of Technology·中国·OpenReview

研究方向

Large Language Models

Diagnosing and Mitigating Systemic Reward Bias in Self-Rewarding RL

ICLR 2026Rejected

PatternKV: Flattening KV Representation Expands Quantization Headroom

ICLR 2026Rejected

合作者 (10)