Fabien Roger

Publication Activity (10 Years)

Years Active: 2022-2024
Publications (10 Years): 7

Top Topics

Language Models For Information Retrieval

Document Ranking

Language Modelling

Top Venues

Trans. Mach. Learn. Res.

Publications

Buck Shlegeris, Fabien Roger, Lawrence Chan, Euan McLean
Language Models Are Better Than Humans at Next-token Prediction. Trans. Mach. Learn. Res. 2024 (2024)
Ryan Greenblatt, Fabien Roger, Dmitrii Krasheninnikov, David Krueger
Stress-Testing Capability Elicitation With Password-Locked Models. CoRR (2024)
Fabien Roger, Ryan Greenblatt, Max Nadeau, Buck Shlegeris, Nate Thomas
Measurement Tampering Detection Benchmark. CoRR (2023)
Fabien Roger, Ryan Greenblatt
Preventing Language Models From Hiding Their Reasoning. CoRR (2023)
Fabien Roger
Large Language Models Sometimes Generate Purely Negatively-Reinforced Text. CoRR (2023)
Ryan Greenblatt, Buck Shlegeris, Kshitij Sachan, Fabien Roger
AI Control: Improving Safety Despite Intentional Subversion. CoRR (2023)
Buck Shlegeris, Fabien Roger, Lawrence Chan, Euan McLean
Language models are better than humans at next-token prediction. CoRR (2022)