factual
Reasoning best from combining test-time and training-time
Reasoning—tasks requiring test-time, step-by-step computation or deduction—is best achieved by combining inference-time computation (the model thinking step by step while deployed) with practice at training time, rather than relying on either approach alone.
factualpending
Evidence Quote
“you'd get the best results by combining these two things.”
Source
John Schulman (OpenAI Cofounder) — Reasoning, RLHF, & plan for 2027 AGI— Dwarkesh Patel PodcastCreated: 6/13/2026, 3:34:30 AM
Your notes
Loading notes...