[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-researchers-teach-ai-to-imagine-what-it-cant-see":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},5515,"researchers-teach-ai-to-imagine-what-it-cant-see","Researchers Teach AI to Imagine What It Can't See","A new technique called Imaginative Perception Tokens helps AI models reason about spaces they can't directly see, often beating chain-of-thought prompting.","A new training technique lets AI models reason about spaces they can't directly see, and it beats the industry's default chain-of-thought approach at its own game.\n\nResearchers introduced Imaginative Perception Tokens, or IPT, a way for vision language models to externalize what they would perceive from an unseen viewpoint or through an occluded space, without generating actual images. They built three benchmark tasks, perspective taking, path tracing, and multiview counting, using about 20,000 examples with ground-truth imagined views included. The team trained the approach on BAGEL, an open vision language model, and found IPT-supervised versions consistently outperformed models trained with standard text chain-of-thought reasoning. On multiview counting, IPT lifted accuracy by 3.4%, and on path tracing it matched strong closed-source models.\n\nThe more telling result is the failure case: forcing spatial reasoning through chain-of-thought text actually degraded performance in this study. That undercuts the common assumption that text-based reasoning chains are a general-purpose fix for any hard problem a model faces.\n\nIt is a narrow academic result, not a product, but it is a reminder that some problems need a different kind of scratchpad, not just a longer one.","[\"ai\",\"spatial reasoning\",\"vision language models\",\"research\"]","2026-08-18T04:00:00.000Z","2026-08-18T23:27:39.179Z","2026-08-18T23:27:51.061Z","published",null,[],"ai",[24,26,27,28],"spatial reasoning","vision language models","research",[30],{"name":31,"url":32},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2606.03988",0,{"sections":35},[36,40,44,49,54,59,64,69,74,78,83,88,93,98],{"name":37,"slug":24,"count":38,"latest_published_at":39},"AI",3293,"2026-08-20T04:00:00.000Z",{"name":41,"slug":42,"count":43,"latest_published_at":39},"Security","security",435,{"name":45,"slug":46,"count":47,"latest_published_at":48},"Policy","policy",210,"2026-08-19T09:32:27.000Z",{"name":50,"slug":51,"count":52,"latest_published_at":53},"Deals","deals",179,"2026-06-29T20:02:07.000Z",{"name":55,"slug":56,"count":57,"latest_published_at":58},"Hardware","hardware",140,"2026-08-19T18:25:42.000Z",{"name":60,"slug":61,"count":62,"latest_published_at":63},"Consumer Tech","consumer-tech",95,"2026-08-18T16:05:00.000Z",{"name":65,"slug":66,"count":67,"latest_published_at":68},"Science","science",90,"2026-08-19T18:41:02.000Z",{"name":70,"slug":71,"count":72,"latest_published_at":73},"Software","software",73,"2026-08-18T07:51:50.000Z",{"name":75,"slug":76,"count":77,"latest_published_at":18},"Dev Tools","dev-tools",69,{"name":79,"slug":80,"count":81,"latest_published_at":82},"Startups","startups",47,"2026-08-19T19:13:46.000Z",{"name":84,"slug":85,"count":86,"latest_published_at":87},"Gaming","gaming",41,"2026-07-09T04:00:00.000Z",{"name":89,"slug":90,"count":91,"latest_published_at":92},"General","general",33,"2026-08-18T22:18:13.000Z",{"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"]