AI can zoom images further than we can annotate them. But without supervision, does it recover details or invent them? <strong>OracleZoom</strong> teaches recursive super-resolution to stay grounded by preserving observable evidence while synthesizing unseen details.</p>\n","updatedAt":"2026-09-10T10:57:58.627Z","author":{"_id":"638940232a897944ea5d2a41","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/638940232a897944ea5d2a41/JSsRSOhkKhuhRKr1CpFB9.jpeg","fullname":"Shubhashis Roy Dipta","name":"dipta007","type":"user","isPro":true,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":19,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.9124044179916382},"editors":["dipta007"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/638940232a897944ea5d2a41/JSsRSOhkKhuhRKr1CpFB9.jpeg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2609.06490","authors":[{"_id":"6aa12d1fff4bf7311191aaa8","user":{"_id":"638940232a897944ea5d2a41","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/638940232a897944ea5d2a41/JSsRSOhkKhuhRKr1CpFB9.jpeg","isPro":true,"fullname":"Shubhashis Roy Dipta","user":"dipta007","type":"user","name":"dipta007"},"name":"Shubhashis Roy Dipta","status":"claimed_verified","statusLastChangedAt":"2026-09-10T00:45:05.045Z","hidden":false},{"_id":"6aa12d1fff4bf7311191aaa9","name":"Sourajit Saha","hidden":false},{"_id":"6aa12d1fff4bf7311191aaaa","name":"Shaswati Saha","hidden":false},{"_id":"6aa12d1fff4bf7311191aaab","name":"Nobin Sarwar","hidden":false}],"mediaUrls":["https://cdn-uploads.huggingface.co/production/uploads/638940232a897944ea5d2a41/So-R0TDExyhMc_tJfDGBE.jpeg","https://cdn-uploads.huggingface.co/production/uploads/638940232a897944ea5d2a41/tFfB7tuV4B3s1LCZaskdo.jpeg"],"publishedAt":"2026-09-06T00:00:00.000Z","submittedOnDailyAt":"2026-09-10T00:00:00.000Z","title":"OracleZoom: On-Policy Self-Distillation Inspired Reference-Constrained Recursive Image Super Resolution","submittedOnDailyBy":{"_id":"638940232a897944ea5d2a41","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/638940232a897944ea5d2a41/JSsRSOhkKhuhRKr1CpFB9.jpeg","isPro":true,"fullname":"Shubhashis Roy Dipta","user":"dipta007","type":"user","name":"dipta007"},"summary":"Recursive Super-Resolution (SR) extends fixed-scale SR to extreme magnification by repeatedly feeding predictions back into the same model, analogous to zooming an image repeatedly. However, ground truth availability at every scale, especially at depth, remains challenging as the required source resolution grows geometrically, leaving deeper predictions unsupervised. We present OracleZoom, an on-policy distillation-inspired, reference-constrained framework that trains on its trajectory while carrying the last ground-truth evidence beyond the supervision boundary. Direct and cross-scale supervision constrain verifiable content, while a no-reference quality objective guides unresolved fine-scale detail. A KL-constrained pretrained latent prior limits quality-driven drift, while EMA consistency stabilizes the supervision boundary. Across seven datasets, OracleZoom achieves the state-of-the-art SR quality across zooming scales, averaging 0.713 CLIPIQA, with larger gains on deeper scales, while significantly reducing hallucinations. Code, data, and models are available at https://dipta007.github.io/OracleZoom/ .","upvotes":4,"discussionId":"6aa12d1fff4bf7311191aaac","projectPage":"https://dipta007.github.io/OracleZoom/","githubRepo":"https://github.com/dipta007/OracleZoom","githubRepoAddedBy":"user","ai_summary":"OracleZoom improves recursive super-resolution by combining trajectory-based training with cross-scale supervision and a latent prior to reduce hallucinations at extreme magnifications.","ai_keywords":["recursive super-resolution","on-policy distillation","cross-scale supervision","no-reference quality objective","KL-constrained pretrained latent prior","EMA consistency"],"ai_summary_model":"thinkingmachines/Inkling-Small","githubStars":0},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"638940232a897944ea5d2a41","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/638940232a897944ea5d2a41/JSsRSOhkKhuhRKr1CpFB9.jpeg","isPro":true,"fullname":"Shubhashis Roy Dipta","user":"dipta007","type":"user"},{"_id":"684d57f26e04c265777ead3f","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/cuOj-bQqukSZreXgUJlfm.png","isPro":false,"fullname":"Joakim Lee","user":"Reinforcement4All","type":"user"},{"_id":"6351e5bb3734c6e8a5c1bec1","avatarUrl":"/avatars/a784a51b369b197398575c3afbd5ceab.svg","isPro":false,"fullname":"Han-Bit Kang","user":"hbkang","type":"user"},{"_id":"6a2da6c8ca070ee12c6e396c","avatarUrl":"/avatars/0355287dcabaa67dbc7f0b10b87451f9.svg","isPro":false,"fullname":"Joe Mama","user":"JoeMama123123123","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2609/2609.06490.md","query":{}}">

Get this paper in your agent:

Don't have the latest CLI?