🌐 Project page: <a href=\"https://duryi.github.io/UA-NWM-Project-Page/\" rel=\"nofollow\">https://duryi.github.io/UA-NWM-Project-Page/</a><br>💻 Code: <a href=\"https://github.com/DurYi/UA-NWM\" rel=\"nofollow\">https://github.com/DurYi/UA-NWM</a><br>🤗 Data: <a href=\"https://huggingface.co/datasets/DurYi/AirGoal-10k\">https://huggingface.co/datasets/DurYi/AirGoal-10k</a><br>📦 Models: <a href=\"https://huggingface.co/DurYi/UA-NWM-Checkpoints\">https://huggingface.co/DurYi/UA-NWM-Checkpoints</a></p>\n","updatedAt":"2026-08-07T09:15:24.415Z","author":{"_id":"67441d04de9997dd26931935","avatarUrl":"/avatars/ae1a6684aaf796a06adea9237001980e.svg","fullname":"Zhu Deyi","name":"DurYi","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.6562063694000244},"editors":["DurYi"],"editorAvatarUrls":["/avatars/ae1a6684aaf796a06adea9237001980e.svg"],"reactions":[],"isReport":false}},{"id":"6a793f1ba421b650a61deb84","author":{"_id":"67441d04de9997dd26931935","avatarUrl":"/avatars/ae1a6684aaf796a06adea9237001980e.svg","fullname":"Zhu Deyi","name":"DurYi","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false},"createdAt":"2026-08-10T03:01:47.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"We present UA-NWM, an uncertainty-aware world model for aerial image-goal navigation. UA-NWM formulates trajectory scoring as conditional OOD detection, models plausible future variations through an uncertainty subspace, and separates plausible uncertainty-induced deviations from unexplained residual errors. This enables efficient distribution-aware trajectory scoring without stochastic future sampling. Experiments show that UA-NWM improves navigation performance across diverse tasks, while preserving low inference latency, and real-world UAV deployment further supports its practical applicability.","html":"<p>We present UA-NWM, an uncertainty-aware world model for aerial image-goal navigation. UA-NWM formulates trajectory scoring as conditional OOD detection, models plausible future variations through an uncertainty subspace, and separates plausible uncertainty-induced deviations from unexplained residual errors. This enables efficient distribution-aware trajectory scoring without stochastic future sampling. Experiments show that UA-NWM improves navigation performance across diverse tasks, while preserving low inference latency, and real-world UAV deployment further supports its practical applicability.</p>\n","updatedAt":"2026-08-10T03:01:47.651Z","author":{"_id":"67441d04de9997dd26931935","avatarUrl":"/avatars/ae1a6684aaf796a06adea9237001980e.svg","fullname":"Zhu Deyi","name":"DurYi","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8479858040809631},"editors":["DurYi"],"editorAvatarUrls":["/avatars/ae1a6684aaf796a06adea9237001980e.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2608.05597","authors":[{"_id":"6a753ef5e1228e04b3238195","user":{"_id":"67441d04de9997dd26931935","avatarUrl":"/avatars/ae1a6684aaf796a06adea9237001980e.svg","isPro":false,"fullname":"Zhu Deyi","user":"DurYi","type":"user","name":"DurYi"},"name":"Deyi Zhu","status":"claimed_verified","statusLastChangedAt":"2026-08-07T08:45:04.530Z","hidden":false},{"_id":"6a753ef5e1228e04b3238196","user":{"_id":"66d70cb5a5098dc770db5a6b","avatarUrl":"/avatars/a742e173a86950e553b7e3696a17c109.svg","isPro":false,"fullname":"Haoyu Fan","user":"s0Meb0dyyyy","type":"user","name":"s0Meb0dyyyy"},"name":"Haoyu Fan","status":"claimed_verified","statusLastChangedAt":"2026-08-07T16:45:28.173Z","hidden":false},{"_id":"6a753ef5e1228e04b3238197","name":"Yinan Zhu","hidden":false},{"_id":"6a753ef5e1228e04b3238198","name":"Weichen Zhang","hidden":false},{"_id":"6a753ef5e1228e04b3238199","name":"Shilin Ma","hidden":false},{"_id":"6a753ef5e1228e04b323819a","name":"Xinlei Chen","hidden":false},{"_id":"6a753ef5e1228e04b323819b","name":"Yansong Tang","hidden":false}],"mediaUrls":["https://cdn-uploads.huggingface.co/production/uploads/67441d04de9997dd26931935/hdmqvdJrjk6-4o92WpW_F.mp4","https://cdn-uploads.huggingface.co/production/uploads/67441d04de9997dd26931935/adzxM1l-SXWnej5SvVqjd.mp4","https://cdn-uploads.huggingface.co/production/uploads/67441d04de9997dd26931935/-1jocZsXOI-VCCU05HdxI.mp4","https://cdn-uploads.huggingface.co/production/uploads/67441d04de9997dd26931935/c9Dd_TaUvaUToez0auWdD.mp4"],"publishedAt":"2026-08-06T00:00:00.000Z","submittedOnDailyAt":"2026-08-10T00:00:00.000Z","title":"Uncertainty-Aware World Model for Aerial Image-Goal Navigation","submittedOnDailyBy":{"_id":"67441d04de9997dd26931935","avatarUrl":"/avatars/ae1a6684aaf796a06adea9237001980e.svg","isPro":false,"fullname":"Zhu Deyi","user":"DurYi","type":"user","name":"DurYi"},"summary":"Aerial image-goal navigation requires an unmanned aerial vehicle (UAV) to reach a target location specified by a goal image. Existing world-model-based methods rank candidate trajectories using predicted futures, but typically rely on only one or a few point predictions, which is inadequate for large-scale outdoor environments with substantial future-state uncertainty. To address this limitation, we propose the Uncertainty-Aware Navigation World Model (UA-NWM), an efficient latent world model for aerial image-goal navigation, which formulates trajectory scoring as conditional out-of-distribution detection. UA-NWM represents plausible futures with an uncertainty subspace and decomposes the prediction--goal discrepancy into uncertainty-explainable and unexplainable components. Only the unexplainable residual is used for scoring, enabling robust selection without multiple future samples. Extensive experiments demonstrate that UA-NWM consistently outperforms existing navigation world models while maintaining low inference latency. Real-world UAV experiments further validate its practical applicability. Project page: https://duryi.github.io/UA-NWM-Project-Page","upvotes":5,"discussionId":"6a753ef6e1228e04b323819c","projectPage":"https://duryi.github.io/UA-NWM-Project-Page","githubRepo":"https://github.com/DurYi/UA-NWM","githubRepoAddedBy":"user","githubStars":2,"organization":{"_id":"628735cbc83a2d6ab8d14a66","name":"Tsinghua","fullname":"Tsinghua University","avatar":"https://www.gravatar.com/avatar/6c5c1441e3283e7543342e59277ea219?d=retro&size=100"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"67441d04de9997dd26931935","avatarUrl":"/avatars/ae1a6684aaf796a06adea9237001980e.svg","isPro":false,"fullname":"Zhu Deyi","user":"DurYi","type":"user"},{"_id":"63ac5701c21e60a3e9b58aa7","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63ac5701c21e60a3e9b58aa7/g6EX7diOpuA94R2ab-rZC.png","isPro":true,"fullname":"Dipankar Sarkar","user":"dipankarsarkar","type":"user"},{"_id":"620783f24e28382272337ba4","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/620783f24e28382272337ba4/zkUveQPNiDfYjgGhuFErj.jpeg","isPro":false,"fullname":"GuoLiangTang","user":"Tommy930","type":"user"},{"_id":"6407e5294edf9f5c4fd32228","avatarUrl":"/avatars/8e2d55460e9fe9c426eb552baf4b2cb0.svg","isPro":false,"fullname":"Stoney Kang","user":"sikang99","type":"user"},{"_id":"66d70cb5a5098dc770db5a6b","avatarUrl":"/avatars/a742e173a86950e553b7e3696a17c109.svg","isPro":false,"fullname":"Haoyu Fan","user":"s0Meb0dyyyy","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"organization":{"_id":"628735cbc83a2d6ab8d14a66","name":"Tsinghua","fullname":"Tsinghua University","avatar":"https://www.gravatar.com/avatar/6c5c1441e3283e7543342e59277ea219?d=retro&size=100"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2608/2608.05597.md","query":{}}">

Get this paper in your agent:

Don't have the latest CLI?