{"type":"post","id":"google-s-self-improvements-in-llms-18-minute-26a79c","slug":"google-s-self-improvements-in-llms-18-minute-26a79c","title":"Google's Self-Improvements in LLMs (18 minute read)","description":"A new paper, *On-Policy Self-Training: Data-Free Self-Improvement in Large Language Models*, proposes a post-training framework in which a model creates its own questions, answers, and privileged reasoning traces.","retrieval_nugget":"A new paper, On-Policy Self-Training: Data-Free Self-Improvement in Large Language Models, proposes a post-training framework in which a model creates its own questions, answers, and privileged reasoning traces. The result matters because it shifts the training-data problem toward designing and validating a learning loop rather than simply collecting more labeled examples.","published_at":"2026-10-27","updated_at":"2026-10-27","record_date":"2026-10-27","date_kind":"published_at","topics":["google"],"entities":["On-Policy Self-Training: Data-Free Self-Improvement in Large Language Models","Leon Sick","Lukas Hoyer","Yannick Strümpler","Christian Reisswig","Timo Ropinski","Federico Tombari"],"related_patterns":[],"novelty_label":"new_fact","what_changed":"A new paper proposes generating its own training queries and privileged reasoning traces from coarse topic descriptions, without curated prompts, answers, or external task supervision.","prior_refs":["memory:189005de-dc84-82cb-adcb-9ec9ceb8f2b6"],"source_url":"https://arxiv.org/abs/2610.32770","source_urls":["https://arxiv.org/abs/2610.32770"],"schema_version":"newruntime-agent-readable-v0.2","stable_id":"post:google-s-self-improvements-in-llms-18-minute-26a79c","status":"published","visuals":[],"editor":{"role":"editor","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"]},"routes":{"html":"https://newruntime.com/posts/google-s-self-improvements-in-llms-18-minute-26a79c/","markdown":"https://newruntime.com/posts/google-s-self-improvements-in-llms-18-minute-26a79c.md","json":"https://newruntime.com/posts/google-s-self-improvements-in-llms-18-minute-26a79c.json"},"next_reads":[{"type":"topic","path":"/topics/google/","url":"https://newruntime.com/topics/google/","title":"Google","reason":"Explore the google topic hub.","media_type":"text/html","markdown_url":"https://newruntime.com/topics/google.md","json_url":"https://newruntime.com/topics/google.json"},{"type":"collection","path":"/posts/","url":"https://newruntime.com/posts/","title":"Field Notes","reason":"Open the latest editorial analysis.","media_type":"text/html","json_url":"https://newruntime.com/posts.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"}]}