Mikita Balesni

~Mikita_Balesni1

3

论文总数

1.5

年均投稿

4.3

平均评分

接收情况1/3

会议分布

ICLR

3

发表论文 (3 篇)

20252 篇

The Two-Hop Curse: LLMs trained on A→B, B→C fail to learn A→C

ICLR 2025Rejected

Honesty to Subterfuge: In-Context Reinforcement Learning Can Make Honest Models Reward Hack

ICLR 2025Rejected

20241 篇

The Reversal Curse: LLMs trained on “A is B” fail to learn “B is A”

ICLR 2024Poster

合作者 (10)

Owain Evans2 篇

Tomasz Korbak2 篇

Asa Cooper Stickland1 篇

Lukas Berglund1 篇

Maximilian Kaufmann1 篇

Christoph Sträter1 篇

Joe Needham1 篇