影响力指数

14.15/100

超过 43.3%

全站排名 #36,506

发表论文2 篇

平均评分6.5

年均产出2.0 篇/年

Nina Rimsky

Researcher@Anthropic·英国·OpenReview

研究方向

Deep learning · Machine learning interpretability · Machine learning

Refusal in Language Models Is Mediated by a Single Direction

NeurIPS 2024Poster

Many-shot Jailbreaking

NeurIPS 2024Poster

合作者 (20)

Ansh Radhakrishnan