影响力指数

-/100

发表论文2 篇

平均评分2.7

年均产出2.0 篇/年

Louis Thomson

Undergrad student@University of Oxford·英国·OpenReview

研究方向

AI safety · LLM evaluations · AI control · multi-agent settings · game theory

Towards shutdownable agents: stochastic choice in unseen gridworlds via DReST rewards

ICLR 2026Rejected

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

ICLR 2026Withdrawn

合作者 (8)

Alessandro Abate

Charlie Griffin

Alexander Roman

Christos Ziakas

Elliott Thornley