影响力指数

32.6/100

前 21.6%

全站排名 #13,879

发表论文4 篇

平均评分4.8

年均产出1.3 篇/年

Luke Marks

Researcher@MATS Research·美国·OpenReview

研究方向

Reinforcement learning · Interpretability · AI safety · Deep learning

Output Supervision Can Obfuscate the Chain of Thought

ICLR 2026Poster

Enhancing Neural Network Interpretability with Feature-Aligned Sparse Autoencoders

ICLR 2025Rejected

合作者 (12)

Alexander Matt Turner