Max Bartolo

~Max_Bartolo1

4

论文总数

2.0

年均投稿

6.5

平均评分

接收情况3/4

会议分布

ICLR

3

NeurIPS

1

发表论文 (4 篇)

20253 篇

Understanding Likelihood Over-optimisation in Direct Alignment Algorithms

ICLR 2025Rejected

Reverse Engineering Human Preferences with Reinforcement Learning

NeurIPS 2025Spotlight

Procedural Knowledge in Pretraining Drives Reasoning in Large Language Models

ICLR 2025Poster

20241 篇

Human Feedback is not Gold Standard

ICLR 2024Poster

合作者 (18)

Acyr Locatelli2 篇

Maximilian Mozes2 篇

Dwaraknath Gnaneshwar1 篇

Edward Grefenstette1 篇

Laura Ruis1 篇

Robert Kirk1 篇

Siddhartha Rao Kamalakara1 篇