QQWorld improves latent world models with a quantile-quantile matching objective that aligns projected latent samples with rank-matched Gaussian quantiles, boosting planning success.</p>\n","updatedAt":"2026-08-03T01:48:52.342Z","author":{"_id":"634e60454677a5891c0902f4","avatarUrl":"/avatars/4dc143719afe7686e05b7f2c2c5c1871.svg","fullname":"Xiangyu Xu","name":"xjcvcvxj","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":1,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.7844775915145874},"editors":["xjcvcvxj"],"editorAvatarUrls":["/avatars/4dc143719afe7686e05b7f2c2c5c1871.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2607.28415","authors":[{"_id":"6a6ff313bbe824e6bcc465a0","name":"Zhoushun Yu","hidden":false},{"_id":"6a6ff313bbe824e6bcc465a1","name":"Xiaoyu Hu","hidden":false},{"_id":"6a6ff313bbe824e6bcc465a2","name":"Xiangyu Xu","hidden":false}],"publishedAt":"2026-07-30T00:00:00.000Z","submittedOnDailyAt":"2026-08-03T00:00:00.000Z","title":"QQWorld: Quantile-Quantile Matching for World Model Regularization","submittedOnDailyBy":{"_id":"634e60454677a5891c0902f4","avatarUrl":"/avatars/4dc143719afe7686e05b7f2c2c5c1871.svg","isPro":false,"fullname":"Xiangyu Xu","user":"xjcvcvxj","type":"user","name":"xjcvcvxj"},"summary":"Latent world models enable efficient planning by predicting future states in a compact representation space, but their performance depends critically on the quality of the learned latent distribution. LeWorldModel (LeWM) regularizes its latents toward an isotropic Gaussian using the Epps-Pulley (EP) objective. We show that the corrective gradients of EP rapidly vanish for isolated tail samples, leaving heavy-tailed deviations insufficiently controlled. To address this limitation, we propose QQWorld, which replaces EP with a quantile-quantile matching objective that directly aligns projected latent samples with rank-matched Gaussian quantiles, thereby maintaining effective corrective gradients in the tails. We further develop cross-batch QQ, which enlarges the effective ranking pool using detached samples from previous batches, and characterize its bias-variance trade-off. Across four control environments, QQWorld effectively improves the average planning success rate of LeWM, while consistently yielding better Gaussian alignment and thinner latent tails.","upvotes":22,"discussionId":"6a6ff314bbe824e6bcc465a3"},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"634e60454677a5891c0902f4","avatarUrl":"/avatars/4dc143719afe7686e05b7f2c2c5c1871.svg","isPro":false,"fullname":"Xiangyu Xu","user":"xjcvcvxj","type":"user"},{"_id":"67652d0f9275b4a80e143c22","avatarUrl":"/avatars/145058c3179cc82ed97703d4c0ed47d5.svg","isPro":false,"fullname":"Yuntian Gao","user":"naiverer","type":"user"},{"_id":"6554a79531bd6430790e2329","avatarUrl":"/avatars/59a15f675c90ea699a372b33564d3977.svg","isPro":false,"fullname":"Jiahuan Zhang","user":"JiahuanZhang","type":"user"},{"_id":"68f38f0465ecaf07402fad2b","avatarUrl":"/avatars/cb5c8fbfe95cbc6d23e4dccd8d4a2612.svg","isPro":false,"fullname":"Rock","user":"f0rest123","type":"user"},{"_id":"67ed0ebc0c63bb84edbf1b58","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/qLL7TfBvr9gUng4iW7sg6.png","isPro":false,"fullname":"Xiewei","user":"Xiewei1211","type":"user"},{"_id":"67bd50d3d50af74904d82dee","avatarUrl":"/avatars/5647abbd909ea664360a1f02094a130f.svg","isPro":false,"fullname":"changjl","user":"chang-j-l","type":"user"},{"_id":"6039478ab3ecf716b1a5fd4d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6039478ab3ecf716b1a5fd4d/_Thy4E7taiSYBLKxEKJbT.jpeg","isPro":true,"fullname":"taesiri","user":"taesiri","type":"user"},{"_id":"67fcd783d1ec7d15ba66575e","avatarUrl":"/avatars/decbcb963b616c789cca71af302d626a.svg","isPro":false,"fullname":"Austine John","user":"AustineJohnBreaker","type":"user"},{"_id":"6703a431c67c24aeada272f6","avatarUrl":"/avatars/5eebd40efdae3783d62f88fe01ee64e8.svg","isPro":false,"fullname":"zsy","user":"whatsZsy","type":"user"},{"_id":"6463554dd2044cd1d7c6e0bf","avatarUrl":"/avatars/d7653623117268c545a7063fec69664b.svg","isPro":false,"fullname":"Bingzheng Wei","user":"Bingzheng","type":"user"},{"_id":"68bd82ec7bcabff61a0c4a97","avatarUrl":"/avatars/457e5384f3dac43175e3e2ccefcb3bcc.svg","isPro":false,"fullname":"YU D Tian","user":"Yufor","type":"user"},{"_id":"6a6a829fa698558ca76157ee","avatarUrl":"/avatars/98e9a42de130397e7f4efce0039cdacd.svg","isPro":false,"fullname":"Richard Williams","user":"dennis-9760525","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"query":{}}">

Cite arxiv.org/abs/2607.28415 in a model README.md to link it from this page.

Cite arxiv.org/abs/2607.28415 in a dataset README.md to link it from this page.

Cite arxiv.org/abs/2607.28415 in a Space README.md to link it from this page.

Add this paper to a collection to link it from this page.