Debate Helps Supervise Unreliable Experts
Paper
• 2311.08702 • Published
• 1
David Rein
Idavidrein
AI & ML interests
safety/alignment, careful data collection projects, LLM evaluations, scalable oversight
Recent Activity
new activity
7 days ago
Idavidrein/gpqa:Set evaluation_framework to inspect-ai in eval.yaml new activity
about 2 months ago
Idavidrein/gpqa:adds_eval_yaml new activity
about 2 months ago
Idavidrein/gpqa:adds_eval_yaml Organizations
None yet
authored 2 papers over 1 year ago