影响力指数

28.56/100

前 26.1%

全站排名 #16,795

发表论文3 篇

平均评分5.7

年均产出1.5 篇/年

Nathan W. Henry

Undergrad student@University of Toronto·加拿大·OpenReview

SteeringSafety: A Systematic Safety Evaluation Framework of Representation Steering in LLMs

ICLR 2026Rejected

RepIt: Steering Language Models with Concept-Specific Refusal Vectors

ICLR 2026Poster

Geometry of Lightning Self-Attention: Identifiability and Dimension

ICLR 2025Poster

合作者 (9)

Nicholas Crispino

Giovanni Luca Marchetti