Shockingly Simple Self-retrospection Improves Agentic Models Without RL | Hacker News - But Better

Shockingly Simple Self-retrospection Improves Agentic Models Without RL arxiv.org

People learn not only by repeating successful actions, but also by recounting and explaining their experiences, revising their understanding to guide future behavior. Can a language-model agent improve its future…

1 points•Betelbuddy•1 day ago•0 comments•

#without #agentic #self #improves #shockingly #retrospection #simple self #shockingly simple

0 comments

No comments yet.

Related stories

MacTube 2 YouTube as a real Mac app – without Shorts and without AI slopgithub.com
Hacker News · 1 points · 5 days ago
Async cancel safety in lilos: Mutex without lock, Queue without pushcliffle.com
Lobsters · 14 points · about 3 years ago
Show HN: Air-gapped file encryption as self-decrypting HTML pagecms-sfx-demo.apeleg.com
Hacker News · 90 points · 7 days ago
A self-evolving agentic system for automated execution of biological protocolsarxiv.org
Hacker News · 1 points · 9 days ago
Show HN: Busbar self-hosted execution boundary for Agentic Appsgithub.com
Hacker News · 1 points · 6 days ago
Halo: An open-source, self-modifiable, agentic operating systemgethalo.dev
Hacker News · 1 points · 3 days ago