[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-new-agent-architecture-splits-reasoning-from-action-checks":10,"sections":35},{"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":30,"feedback":34,"feedback_at":22,"cost_usd":34,"total_tokens":34},9865,"new-agent-architecture-splits-reasoning-from-action-checks","New Agent Architecture Splits Reasoning From Action Checks","A new arXiv study adds separate action-checking and completion-checking roles to AI agents, and the clearest gains go to weaker models, not top-tier ones.","DeReAct gives AI agents a built-in skeptic instead of letting one model grade its own homework.\n\nRight now, most ReAct-style agents use a single language model to propose an action, carry it out, and decide when the task is done, which lets early mistakes snowball into false claims of success. DeReAct, described in a new arXiv paper, splits that job in two: a Critic checks a proposed action before it executes, and a Context Manager tracks what the environment has actually confirmed and certifies when a task is truly complete. Tested on the GAIA and SWE-bench Verified benchmarks, the setup lifted Pass@1 scores by 6.5 to 7.0 points for Qwen3-Coder-480B and by 4.2 to 5.2 points for Claude Sonnet 4.5.\n\nThose numbers matter less than the pattern behind them: the benefit shrinks as the underlying model gets better. With Claude Opus 4.5, Pass@1 barely moved, though the agent still produced more evidence-backed trajectories with fewer unsupported completion claims. That suggests external oversight layers are compensating for weaknesses that stronger models have already learned to avoid on their own.\n\nWorth remembering next time a benchmark post calls extra scaffolding a universal fix: sometimes it is just a patch for a model that was not that reliable to begin with.","[\"ai-agents\",\"llm-benchmarks\",\"arxiv-research\",\"agent-architecture\"]","2026-10-05T04:00:00.000Z","2026-10-05T11:04:50.128Z","2026-10-05T11:04:56.144Z","published",null,[],"ai",[26,27,28,29],"ai-agents","llm-benchmarks","arxiv-research","agent-architecture",[31],{"name":32,"url":33},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2610.02351",0,{"sections":36},[37,40,44,49,54,59,63,68,72,76,81,86,91,96],{"name":38,"slug":24,"count":39,"latest_published_at":18},"AI",6165,{"name":41,"slug":42,"count":43,"latest_published_at":18},"Security","security",859,{"name":45,"slug":46,"count":47,"latest_published_at":48},"Policy","policy",444,"2026-10-03T15:02:01.000Z",{"name":50,"slug":51,"count":52,"latest_published_at":53},"Deals","deals",323,"2026-10-04T13:00:00.000Z",{"name":55,"slug":56,"count":57,"latest_published_at":58},"Hardware","hardware",204,"2026-10-03T14:50:50.000Z",{"name":60,"slug":61,"count":62,"latest_published_at":18},"Science","science",177,{"name":64,"slug":65,"count":66,"latest_published_at":67},"Consumer Tech","consumer-tech",158,"2026-10-03T03:21:12.000Z",{"name":69,"slug":70,"count":71,"latest_published_at":18},"Dev Tools","dev-tools",97,{"name":73,"slug":74,"count":71,"latest_published_at":75},"Software","software","2026-10-04T10:00:00.000Z",{"name":77,"slug":78,"count":79,"latest_published_at":80},"Startups","startups",92,"2026-10-04T14:36:25.000Z",{"name":82,"slug":83,"count":84,"latest_published_at":85},"Gaming","gaming",53,"2026-10-02T02:50:39.000Z",{"name":87,"slug":88,"count":89,"latest_published_at":90},"General","general",51,"2026-10-05T02:35:01.000Z",{"name":92,"slug":93,"count":94,"latest_published_at":95},"Reviews","reviews",32,"2026-10-02T18:00:00.000Z",{"name":97,"slug":98,"count":99,"latest_published_at":100},"How-To","how-to",7,"2026-10-01T09:00:00.000Z"]