SPoT paper "Self-Improving Training & Thinking" now on arXiv with blog and thread
polkirichenko · x · 2026-11-12
Final post of the SPoT thread collects all resources: the paper (arXiv 2601.21343, with Chapter 3 covering post-thinking), the official blog, and collaborator Jason Weston's thread. The 14-author team includes Ellen Xiaoqing Tan, Jack Lanchantin, Polina Kirichenko, Jason Weston, and Olga Golovneva.
The paper argues that the classic pretraining/post-training split limits safety, factuality, and reasoning behaviors to late stages; the work introduces thinking and reinforcement earlier by using a strong post-trained model to rewrite data and judge rollouts.
Related event: Meta's SPoT: self-improving post-thinking training(4 posts)→
More from Research
- As models get stronger, the next trainable tasks look increasingly like alignment itself — herbiebradley · 2026-11-12
- RoboCap-10k: 10,000 hours of open egocentric robot data across 21 industries released — micoolcho · 2026-11-12
- Talent vs. luck model reproduced: equal funding is ~9x more efficient than funding winners — ChenhaoTan · 2026-11-12
- Stanford HAI dashboards show employment in AI-exposed jobs down over 4% this year — StanfordHAI · 2026-11-12
- Shun-ichi Amari on how 'do as you like' freedom birthed the math behind AI's 60 years — FrnkNlsn · 2026-11-12
- Formal verification proves a 3,000-step Sokoban optimum, with a bounty on beating it — banteg · 2026-11-12