影响力指数

29.92/100

前 24.5%

全站排名 #15,775

发表论文3 篇

平均评分5.4

年均产出1.5 篇/年

Yang Xu

Undergrad student@Zhejiang University·中国·OpenReview

研究方向

Mechanistic Interpretability · Deep Learning Theory

LH-DECEPTION: Simulating and Understanding LLM Deceptive Behaviors in Long-Horizon Interactions

ICLR 2026Poster

Tracking the Feature Dynamics in LLM Training: A Mechanistic Study

ICLR 2026Rejected

Towards the Dynamics of a DNN Learning Symbolic Interactions

NeurIPS 2024Poster

合作者 (14)