[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"topic-en-prerl-training-llms-in-pre-train-space":3},{"cluster":4,"timeline":16},{"id":5,"slug":6,"title":7,"pinned":8,"status":9,"summary":10,"category":11,"language":12,"created_at":13,"merged_into":14,"article_count":15,"first_seen_at":13,"last_updated_at":13},"fae5809e-f43d-4b4b-8c37-fbd3c2829eaf","prerl-training-llms-in-pre-train-space","PreRL: Training LLMs in pre-train space",false,"active","PreRL shifts reinforcement learning from P(y|x) to P(y), using reward-driven updates in pre-train space to improve reasoning and exploration.","research","en","2026-07-05T09:04:02.760914+00:00",null,1,[17],{"id":18,"slug":19,"title":7,"summary":10,"category":11,"image_url":20,"cover_image":20,"published_at":21,"is_canonical_seed":22},"d1bbd868-15d4-459c-9e2b-2626c779b4ef","prerl-training-llms-in-pre-train-space-en","https:\u002F\u002Fxxdpdyhzhpamafnrdkyq.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Fcovers\u002Finline-1776319621187-aig1.png","2026-04-16T06:06:38.221+00:00",true]