PDR has been accepted at ICML 2026! 🇰🇷
Reasoning doesn’t have to mean longer chains of thought:
PDR = draft in parallel → distill into a compact workspace → refine, and shift the Pareto frontier.
arxiv.org/abs/2510.01123
John Quan retweeted
(1/9) Experience replay can cut LLM RL training compute by up to ~40% (without hurting final accuracy—and sometimes improving it).
Paper: arxiv.org/abs/2604.08706
John Quan retweeted
1/ #1stProof. Our second installment — this time tackling Problem 3, with @scottnarmstrong and @MunosRemi
Also check out our takeaways — and a short “Humor from your bot” interlude — below.







