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
52.03/100
前 7.6%
全站排名 #4,874
发表论文5 篇
平均评分
年均产出2.5 篇/年
Haolin Yang
研究方向
natural language processing · large language models · in-context learning · mechanistic interpretability
17
Mechanism of Task-oriented Information Removal in In-context Learning
ICLR 2026Poster
二作20
Localizing Task Recognition and Task Learning in In-Context Learning via Attention Head Analysis
ICLR 2026Poster
一作14
Task Vectors, Learned Not Extracted: Performance Gains and Mechanistic Insights
ICLR 2026Poster
一作7
Binary Autoencoder for Mechanistic Interpretability of Large Language Models
ICLR 2026Rejected
二作