[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-boston-dynamics-spot-learns-to-resolve-vague-commands-on-its-own":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},8338,"boston-dynamics-spot-learns-to-resolve-vague-commands-on-its-own","Boston Dynamics Spot Learns to Resolve Vague Commands on Its Own","CLUE pairs an LLM with a live map so a Spot robot can explore, test hypotheses, and figure out vague commands instead of guessing wrong.","Researchers have built a system that lets a robot dog figure out what you actually meant, without asking you to clarify.\n\nThe framework, called CLUE (Closed-Loop contextual Uncertainty rEsolution), tackles a real gap in robot language models: most assume a task is fully specified and that a map of the environment already exists. CLUE instead uses a large language model to guess at what a vague command might require, then builds a language-embedded map of the space in real time as the robot explores. It tests those guesses through closed-loop interaction, revising its plan as it gathers new information rather than committing to one interpretation upfront. The team ran it on a Boston Dynamics Spot across three indoor and outdoor sites, covering 15 tasks that involved disambiguating objects, inferring what things are used for, and reasoning around occlusion.\n\nThe results suggest the exploration step is doing real work: CLUE landed within 7 percentage points of an oracle policy that already knows the answer, and beat a non-interactive LLM planner by a 4x margin. A separate test showed that just building a smarter map and querying it, without the closed loop, gets roughly a third of CLUE's success rate while burning through more than 10 times as many vision-language-model tokens. That's a pointed rebuttal to the assumption that better maps alone solve ambiguity - the active testing loop appears to be the actual fix.\n\nWorth remembering this is a research paper, not a shipping product - Spot still needed real time in real hallways to sort out what 'the thing by the window' meant, and that is a slower, costlier process than a demo video ever makes it look.","[\"robotics\",\"language models\",\"boston dynamics\",\"research\"]","2026-09-28T04:00:00.000Z","2026-09-29T00:57:48.958Z","2026-09-29T00:57:56.378Z","published",null,[],"ai",[26,27,28,29],"robotics","language models","boston dynamics","research",[31],{"name":32,"url":33},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2609.30428",0,{"sections":36},[37,41,46,51,56,61,66,71,76,81,86,91,96,101],{"name":38,"slug":24,"count":39,"latest_published_at":40},"AI",4917,"2026-09-28T23:39:20.000Z",{"name":42,"slug":43,"count":44,"latest_published_at":45},"Security","security",768,"2026-09-29T01:20:00.000Z",{"name":47,"slug":48,"count":49,"latest_published_at":50},"Policy","policy",406,"2026-09-28T19:04:09.000Z",{"name":52,"slug":53,"count":54,"latest_published_at":55},"Deals","deals",272,"2026-09-28T17:42:46.000Z",{"name":57,"slug":58,"count":59,"latest_published_at":60},"Hardware","hardware",191,"2026-09-28T15:45:00.000Z",{"name":62,"slug":63,"count":64,"latest_published_at":65},"Science","science",154,"2026-09-28T13:19:18.000Z",{"name":67,"slug":68,"count":69,"latest_published_at":70},"Consumer Tech","consumer-tech",139,"2026-09-28T17:09:47.000Z",{"name":72,"slug":73,"count":74,"latest_published_at":75},"Software","software",91,"2026-09-25T20:55:00.000Z",{"name":77,"slug":78,"count":79,"latest_published_at":80},"Dev Tools","dev-tools",87,"2026-09-28T16:11:42.000Z",{"name":82,"slug":83,"count":84,"latest_published_at":85},"Startups","startups",80,"2026-09-28T17:50:28.000Z",{"name":87,"slug":88,"count":89,"latest_published_at":90},"General","general",49,"2026-09-28T16:44:57.000Z",{"name":92,"slug":93,"count":94,"latest_published_at":95},"Gaming","gaming",48,"2026-09-25T18:35:21.000Z",{"name":97,"slug":98,"count":99,"latest_published_at":100},"Reviews","reviews",31,"2026-09-28T14:31:34.000Z",{"name":102,"slug":103,"count":104,"latest_published_at":105},"How-To","how-to",6,"2026-06-16T09:00:00.000Z"]