{"type":"analysis","id":"analysis:research-rlmf-teaches-small-open-models-to-state-confidence-that-matches-their-sampling","slug":"research-rlmf-teaches-small-open-models-to-state-confidence-that-matches-their-sampling","title":"RLMF teaches small open models to state confidence that matches their sampling consistency, but the result is narrow and so far unreplicated","description":"A Yale/Google Research paper accepted at COLM 2026 rescales GRPO advantages by how well a model predicts its own calibration performance. On 8B models it reports large gains in \"faithful calibration.\" The headline \"up to 63%\" is a relative gain over the authors' own RL baseline on one model, not a broad capability jump.","published_at":"2026-10-24T18:00:00.000Z","updated_at":"2026-10-24T18:00:00.000Z","record_date":"2026-10-24","topics":["gemini","prompting","research"],"source_urls":["https://arxiv.org/html/2606.32032v2","https://arxiv.org/abs/2606.32032","https://github.com/yale-nlp/RLMF","https://pybeebee.com/publication/26-rlmf/"],"language":"en","routes":{"html":"https://newruntime.com/analysis/research-rlmf-teaches-small-open-models-to-state-confidence-that-matches-their-sampling/","markdown":"https://newruntime.com/analysis/research-rlmf-teaches-small-open-models-to-state-confidence-that-matches-their-sampling.md","json":"https://newruntime.com/analysis/research-rlmf-teaches-small-open-models-to-state-confidence-that-matches-their-sampling.json"},"author":{"role":"author","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]},"editorial_provenance":{"desk":"authorial","target_ref":"deep_research:953c6cc0-2090-41d7-a999-44bbc5ee6d14","authorial_article_id":"953c6cc0-2090-41d7-a999-44bbc5ee6d14"},"next_reads":[{"type":"topic","path":"/topics/gemini/","url":"https://newruntime.com/topics/gemini/","title":"Gemini","reason":"Explore the gemini topic hub.","media_type":"text/html","markdown_url":"https://newruntime.com/topics/gemini.md","json_url":"https://newruntime.com/topics/gemini.json"},{"type":"topic","path":"/topics/prompting/","url":"https://newruntime.com/topics/prompting/","title":"Prompting","reason":"Explore the prompting topic hub.","media_type":"text/html","markdown_url":"https://newruntime.com/topics/prompting.md","json_url":"https://newruntime.com/topics/prompting.json"},{"type":"topic","path":"/topics/research/","url":"https://newruntime.com/topics/research/","title":"Research","reason":"Explore the research topic hub.","media_type":"text/html","markdown_url":"https://newruntime.com/topics/research.md","json_url":"https://newruntime.com/topics/research.json"},{"type":"collection","path":"/analysis/","url":"https://newruntime.com/analysis/","title":"Analysis","reason":"Open the latest authorial analysis.","media_type":"text/html","json_url":"https://newruntime.com/analysis.json"},{"type":"collection","path":"/source-ledger.json","url":"https://newruntime.com/source-ledger.json","title":"Source Ledger","reason":"Inspect the public source evidence graph.","media_type":"application/json"}]}