[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-new-test-time-method-makes-diffusion-models-pick-better-images":10,"sections":41},{"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":30,"tags":31,"sources":36,"feedback":40,"feedback_at":22,"cost_usd":40,"total_tokens":40},10712,"new-test-time-method-makes-diffusion-models-pick-better-images","New Test-Time Method Makes Diffusion Models Pick Better Images","BoNG steers diffusion models toward higher-reward images during generation, not just at selection, beating rivals in most tests while running faster.","A new technique called Best-of-N Guidance nudges AI image generators toward better outputs while they are still generating, not just after the fact.\n\nDiffusion models are often steered with a reward model after the fact: generate a batch of N images, score them, and keep the best one. That is Best-of-N sampling, and it only acts at the finish line - it does not change how the earlier denoising steps unfold, so it mostly helps single best-pick outputs rather than the overall batch. The new method, called Best-of-N Guidance (BoNG), builds reward selection into the denoising process itself. It runs online best-of-N selection across a population of denoising particles and uses the current frontrunner as a guidance signal for the rest, pushing the whole batch toward higher-reward regions as it generates.\n\nThat matters because it fixes a real limitation of Best-of-N: it improves the average output, not just the cherry-picked one, which is the difference between a demo trick and a method useful when you need several good images at once. In 36 head-to-head tests against Best-of-N and a particle-based rival called SMC, BoNG won 29 of them, about 81 percent, and against the leading sample-based guidance method it hit a 1.3x higher ImageReward score while running 1.6x faster.\n\nReward-guided generation has quietly become the workhorse behind a lot of 'aligned' AI image output; BoNG's pitch is less a new idea than a sharper version of one everyone already leans on, worth watching once someone runs it outside a benchmark table.","[\"diffusion-models\",\"ai-research\",\"image-generation\",\"test-time-alignment\"]","2026-10-07T04:00:00.000Z","2026-10-09T05:50:38.737Z","2026-10-09T05:50:42.757Z","published",null,[24],{"id":25,"reviewer":26,"round":27,"reason":28,"status":29},"editor-r1","editor",1,"The source states BoNG achieves a 1.3x higher ImageReward score than the rival guidance method (at 1.6x the speed), but the article incorrectly says it 'matched' that method's score at 1.6x speed — fix this to reflect the actual reported performance gain, not parity.","resolved","ai",[32,33,34,35],"diffusion-models","ai-research","image-generation","test-time-alignment",[37],{"name":38,"url":39},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2610.05108",0,{"sections":42},[43,47,52,57,62,67,71,76,81,85,90,95,100,105],{"name":44,"slug":30,"count":45,"latest_published_at":46},"AI",6517,"2026-10-08T04:00:00.000Z",{"name":48,"slug":49,"count":50,"latest_published_at":51},"Security","security",911,"2026-10-07T19:53:42.000Z",{"name":53,"slug":54,"count":55,"latest_published_at":56},"Policy","policy",474,"2026-10-07T18:23:21.000Z",{"name":58,"slug":59,"count":60,"latest_published_at":61},"Deals","deals",458,"2026-10-08T10:03:39.000Z",{"name":63,"slug":64,"count":65,"latest_published_at":66},"Hardware","hardware",223,"2026-10-08T08:30:04.000Z",{"name":68,"slug":69,"count":70,"latest_published_at":18},"Science","science",187,{"name":72,"slug":73,"count":74,"latest_published_at":75},"Consumer Tech","consumer-tech",175,"2026-10-08T09:16:05.000Z",{"name":77,"slug":78,"count":79,"latest_published_at":80},"Software","software",113,"2026-10-07T18:10:00.000Z",{"name":82,"slug":83,"count":79,"latest_published_at":84},"Startups","startups","2026-10-07T23:36:57.000Z",{"name":86,"slug":87,"count":88,"latest_published_at":89},"Dev Tools","dev-tools",105,"2026-10-07T16:59:11.000Z",{"name":91,"slug":92,"count":93,"latest_published_at":94},"General","general",61,"2026-10-07T22:00:24.000Z",{"name":96,"slug":97,"count":98,"latest_published_at":99},"Gaming","gaming",56,"2026-10-07T12:00:00.000Z",{"name":101,"slug":102,"count":103,"latest_published_at":104},"Reviews","reviews",33,"2026-10-05T11:57:17.000Z",{"name":106,"slug":107,"count":108,"latest_published_at":109},"How-To","how-to",8,"2026-10-05T09:00:00.000Z"]