[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-rl-training-teaches-ai-to-combine-skills-not-just-recall-them":10,"sections":34},{"siteName":4,"siteTagline":5,"publisherName":4,"contactEmail":6},"The Revision","Tech news, decoded.","editor@therevision.news",{"gaMeasurementId":8,"adsenseClientId":9},"G-ZW2MV82GYR","ca-pub-8533917693782264",{"article":11},{"id":12,"slug":13,"title":14,"dek":15,"body_md":16,"tags_json":17,"published_at":18,"created_at":19,"updated_at":20,"status":21,"review_note":22,"review_notes":23,"image_url":22,"persona_id":22,"persona_name":22,"section":24,"tags":25,"sources":29,"feedback":33,"feedback_at":22,"cost_usd":33,"total_tokens":33},4475,"rl-training-teaches-ai-to-combine-skills-not-just-recall-them","RL Training Teaches AI to Combine Skills, Not Just Recall Them","New research shows reinforcement learning builds genuinely new reasoning strategies from primitive skills, not just surfacing what the base model already knew.","Reinforcement learning post-training doesn't just unlock latent abilities in AI models — it actively builds new ones by combining simpler skills.\n\nResearchers tested a Transformer pretrained on basic symbol-rewrite operations, then post-trained it using reinforcement learning on a more complex reasoning task with only a binary right-or-wrong reward signal. RL solved problems the base model rarely cracked even with far more attempts. Trace analysis revealed a two-phase mechanism: the model first sharpens its primitive operations, then assembles them into new composite procedures — collapsing ordered chains into single steps or combining independent operations in parallel. Crucially, these aren't one-off tricks; the model reuses and consolidates them into a stable repertoire.\n\nThe finding cuts against the common framing that RL post-training is just \"eliciting\" capabilities already baked into pretraining. The study also pinpoints what makes RL outperform rejection fine-tuning: not more exploration, but more selective exploration. Rejection fine-tuning produces a flood of shortcut-like rewrites, many of them invalid; RL concentrates its search on valid, reusable structure.\n\nFor labs betting on post-training to squeeze more reasoning out of existing models, the implication is double-edged — RL can genuinely compose new strategies, but only if pretraining already organizes primitive skills into reduction procedures that RL has something to compress. Weak foundations don't get papered over; they just produce a weaker ceiling.","[\"ai\",\"machine-learning\",\"reinforcement-learning\",\"research\"]","2026-07-09T04:00:00.000Z","2026-07-09T04:56:06.727Z","2026-07-09T04:56:09.596Z","published",null,[],"ai",[24,26,27,28],"machine-learning","reinforcement-learning","research",[30],{"name":31,"url":32},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2607.07646",0,{"sections":35},[36,40,45,50,55,60,65,70,75,80,84,88,93,98],{"name":37,"slug":24,"count":38,"latest_published_at":39},"AI",2590,"2026-07-16T04:00:00.000Z",{"name":41,"slug":42,"count":43,"latest_published_at":44},"Security","security",294,"2026-07-15T19:59:48.000Z",{"name":46,"slug":47,"count":48,"latest_published_at":49},"Deals","deals",179,"2026-06-29T20:02:07.000Z",{"name":51,"slug":52,"count":53,"latest_published_at":54},"Policy","policy",158,"2026-07-16T00:02:48.000Z",{"name":56,"slug":57,"count":58,"latest_published_at":59},"Hardware","hardware",122,"2026-07-14T19:46:26.000Z",{"name":61,"slug":62,"count":63,"latest_published_at":64},"Consumer Tech","consumer-tech",93,"2026-07-13T13:20:48.000Z",{"name":66,"slug":67,"count":68,"latest_published_at":69},"Software","software",70,"2026-07-13T19:52:25.000Z",{"name":71,"slug":72,"count":73,"latest_published_at":74},"Science","science",66,"2026-07-10T10:29:37.000Z",{"name":76,"slug":77,"count":78,"latest_published_at":79},"Dev Tools","dev-tools",59,"2026-07-07T04:00:00.000Z",{"name":81,"slug":82,"count":83,"latest_published_at":18},"Gaming","gaming",41,{"name":85,"slug":86,"count":83,"latest_published_at":87},"Startups","startups","2026-06-29T20:55:50.000Z",{"name":89,"slug":90,"count":91,"latest_published_at":92},"General","general",29,"2026-07-10T22:28:58.000Z",{"name":94,"slug":95,"count":96,"latest_published_at":97},"Reviews","reviews",20,"2026-06-24T12:00:01.000Z",{"name":99,"slug":100,"count":101,"latest_published_at":102},"How-To","how-to",6,"2026-06-16T09:00:00.000Z"]