影响力指数

15.45/100

超过 46.8%

全站排名 #34,217

发表论文4 篇

平均评分5.8

年均产出2.0 篇/年

Charlie Griffin

PhD student@Department of Computer Science·英国·OpenReview

研究方向

Reward Design in Reinforcement Learning

Subversion Strategy Eval: Can LLMs statelessly strategize to subvert control protocols?

ICLR 2026Rejected

Games for AI Control: Models of Safety Evaluations of AI Deployment Protocols

ICLR 2026Withdrawn

On the Expressivity of Objective-Specification Formalisms in Reinforcement Learning

ICLR 2024Poster

Goodhart's Law in Reinforcement Learning

ICLR 2024Poster

合作者 (14)

Alessandro Abate

Joar Max Viktor Skalse

Alex Troy Mallen

Jacek Karwowski

Klaus Kiendlhofer