[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-llm-agents-run-robots-without-training-data":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},2903,"llm-agents-run-robots-without-training-data","LLM Agents Run Robots Without Training Data","A new framework called FAEA uses a general-purpose LLM agent to control robot arms at near-demo-trained accuracy, no task demonstrations required.","A research team has shown that the same agent framework used to write software can direct a robot arm with no task-specific training data.\n\nThe system, called FAEA (Frontier Agent as Embodied Agent), plugs an unmodified LLM agent framework directly into robotic manipulation benchmarks. The researchers tested it using the Claude Agent SDK across three standard benchmarks — LIBERO, ManiSkill3, and MetaWorld — achieving success rates of 84.9%, 85.7%, and 96% respectively when the agent had access to full environment state. Add one round of human feedback on LIBERO and the number ticks up to 88.2%. Those figures are close to what vision-language-action models reach when trained on fewer than 100 demonstrations per task.\n\nThe gap that matters here is the zero-shot starting point. Most robotic control research still assumes you can collect task-specific demonstrations and fine-tune a model on them — a process that is expensive, brittle under domain shift, and has to restart whenever the task changes. FAEA sidesteps that by leaning on the iterative reasoning that LLM agents already use to debug code, applying it instead to physical manipulation strategy. The practical upside the authors flag is that the system can explore new simulation environments on its own and produce training trajectories for other models.\n\nThe obvious caveat: \"privileged environment state access\" is not the real world. These benchmarks hand the agent clean structured data about object positions and joint states — inputs that a real robot would have to perceive imperfectly through sensors. Still, if frontier model capability keeps compounding, a framework that needs no retraining to benefit from the next model release is a more durable infrastructure bet than one locked to a fine-tuned checkpoint.","[\"robotics\",\"llm\",\"ai-agents\",\"embodied-ai\"]","2026-06-30T04:00:00.000Z","2026-06-30T14:49:50.467Z","2026-06-30T14:49:53.447Z","published",null,[],"ai",[26,27,28,29],"robotics","llm","ai-agents","embodied-ai",[31],{"name":32,"url":33},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2601.20334",0,{"sections":36},[37,41,46,51,56,61,66,71,76,81,86,90,95,100],{"name":38,"slug":24,"count":39,"latest_published_at":40},"AI",2590,"2026-07-16T04:00:00.000Z",{"name":42,"slug":43,"count":44,"latest_published_at":45},"Security","security",294,"2026-07-15T19:59:48.000Z",{"name":47,"slug":48,"count":49,"latest_published_at":50},"Deals","deals",179,"2026-06-29T20:02:07.000Z",{"name":52,"slug":53,"count":54,"latest_published_at":55},"Policy","policy",158,"2026-07-16T00:02:48.000Z",{"name":57,"slug":58,"count":59,"latest_published_at":60},"Hardware","hardware",122,"2026-07-14T19:46:26.000Z",{"name":62,"slug":63,"count":64,"latest_published_at":65},"Consumer Tech","consumer-tech",93,"2026-07-13T13:20:48.000Z",{"name":67,"slug":68,"count":69,"latest_published_at":70},"Software","software",70,"2026-07-13T19:52:25.000Z",{"name":72,"slug":73,"count":74,"latest_published_at":75},"Science","science",66,"2026-07-10T10:29:37.000Z",{"name":77,"slug":78,"count":79,"latest_published_at":80},"Dev Tools","dev-tools",59,"2026-07-07T04:00:00.000Z",{"name":82,"slug":83,"count":84,"latest_published_at":85},"Gaming","gaming",41,"2026-07-09T04:00:00.000Z",{"name":87,"slug":88,"count":84,"latest_published_at":89},"Startups","startups","2026-06-29T20:55:50.000Z",{"name":91,"slug":92,"count":93,"latest_published_at":94},"General","general",29,"2026-07-10T22:28:58.000Z",{"name":96,"slug":97,"count":98,"latest_published_at":99},"Reviews","reviews",20,"2026-06-24T12:00:01.000Z",{"name":101,"slug":102,"count":103,"latest_published_at":104},"How-To","how-to",6,"2026-06-16T09:00:00.000Z"]