影响力指数

-/100

发表论文2 篇

平均评分5.0

年均产出2.0 篇/年

Joris Postmus

Undergrad student@University of Groningen·荷兰·OpenReview

研究方向

mechanistic interpretability · representation engineering · activation steering · large language models · deep learning

From Steering Vectors to Conceptors: Compositional Affine Activation Steering for LLMs

NeurIPS 2025Rejected

From Steering Vectors to Conceptors and Beyond: Compositional Affine Steering Mechanisms for LLMs

ICLR 2025Withdrawn

合作者 (7)

Alexander Müller

Alice S. Dauphin

Guillaume Pourcel

Ilija Lichkovski

Jeremias Lino Ferrao

Kurt Felix Michalak