causal

Generalization from verifiable RL tasks to hard problems

Easier verifiable RL training problems generalized strongly enough that the model could contribute to a very hard problem like a Millennium Prize Problem, even though training likely never involved problems that ambitious, and Brown agrees the gap exists but says training on some hard tasks has shown more ambitious abilities.

causalpending

Speaker

Noam Brown

Evidence Quote

“We see that if we train on some kinds of tasks, it’s able to do tasks that are more ambitious than that.”

Source

Dwarkesh Patel Podcast - Noam Brown – Agent swarms, alignment, & recursive self-improvement— Dwarkesh Patel Podcast
Created: 11/17/2026, 6:12:01 AM

Your notes

Loading notes...