Kshitij Sachan

2 papers · 2024–2024 · 1 conference · across top CS/AI conferences

Conferences

ICML (2)

Top co-authors

Ethan Perez (1) Dan Valentine (1) Edward Grefenstette (1) Fabien Roger (1) Samuel R. Bowman (1) Ryan Greenblatt (1) Akbir Khan (1) Ansh Radhakrishnan (1) Tim Rocktäschel (1) Buck Shlegeris (1)

Papers

AI Control: Improving Safety Despite Intentional Subversion ICML 2024

Debating with More Persuasive LLMs Leads to More Truthful Answers ICML 2024