Project page: <a href=\"https://spyzixuan.github.io/FedOT/\" rel=\"nofollow\">https://spyzixuan.github.io/FedOT/</a></p>\n","updatedAt":"2026-06-24T02:52:22.547Z","author":{"_id":"646c77911ee398a4e9404b8b","avatarUrl":"/avatars/05d1ea421dd4f3e2fd47cbe99fc52933.svg","fullname":"Yunqiu Xu","name":"Yunqiu","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.4977530241012573},"editors":["Yunqiu"],"editorAvatarUrls":["/avatars/05d1ea421dd4f3e2fd47cbe99fc52933.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2606.22875","authors":[{"_id":"6a3a0f6afdcd3514343bb688","name":"Wenlong Cheng","hidden":false},{"_id":"6a3a0f6afdcd3514343bb689","name":"Yuan Gan","hidden":false},{"_id":"6a3a0f6afdcd3514343bb68a","user":{"_id":"646c77911ee398a4e9404b8b","avatarUrl":"/avatars/05d1ea421dd4f3e2fd47cbe99fc52933.svg","isPro":false,"fullname":"Yunqiu Xu","user":"Yunqiu","type":"user","name":"Yunqiu"},"name":"Yunqiu Xu","status":"claimed_verified","statusLastChangedAt":"2026-06-23T13:56:10.517Z","hidden":false},{"_id":"6a3a0f6afdcd3514343bb68b","name":"Jiaxu Miao","hidden":false}],"publishedAt":"2026-06-22T00:00:00.000Z","submittedOnDailyAt":"2026-06-24T00:00:00.000Z","title":"FedOT: Ownership Verification and Leakage Tracing via Watermarks for Federated LDMs","submittedOnDailyBy":{"_id":"646c77911ee398a4e9404b8b","avatarUrl":"/avatars/05d1ea421dd4f3e2fd47cbe99fc52933.svg","isPro":false,"fullname":"Yunqiu Xu","user":"Yunqiu","type":"user","name":"Yunqiu"},"summary":"Training Latent Diffusion Models (LDMs) within Federated Learning (FL) has attracted increasing attention due to its ability to combine the powerful generative capacity of LDMs with the privacy-preserving properties of FL. However, FL requires sharing the global model with multiple participants, which risks unauthorized model distribution or resale by malicious clients. While an intuitive approach is to adopt existing VAE-based watermarking techniques for LDMs in FL, this strategy falls short in addressing such threats due to two fundamental challenges: (1) Existing methods support ownership verification but lack the ability to trace model leakage to a specific malicious client; (2) VAE-based watermarks are vulnerable, as they can be removed simply by replacing the decoder with a clean counterpart. In this paper, we propose FedOT, the first framework for ownership verification and leakage tracing in federated LDMs. Specifically, to address the first challenge, we design a chunked watermark, where the first part is for ownership verification, and the second part is used for client identification. Furthermore, to overcome the second challenge and secure the model against VAE replacement attack, we introduce Latent Vector Transformation (LVT), which strengthens the connection between the VAE and U-Net latent spaces by modifying the original latent distribution of the VAE. Consequently, any attempt to replace the VAE for watermark removal leads to significant image quality degradation, making the LDM model unusable. Extensive experiments demonstrate that FedOT achieves superior performance in both ownership verification and traceability. Project page: https://spyzixuan.github.io/FedOT/.","upvotes":9,"discussionId":"6a3a0f6bfdcd3514343bb68c","projectPage":"https://spyzixuan.github.io/FedOT","ai_summary":"FedOT is a novel framework that enables ownership verification and leakage tracing in federated latent diffusion models by introducing chunked watermarking and latent vector transformation to prevent watermark removal attacks.","ai_keywords":["latent diffusion models","federated learning","watermarking","ownership verification","leakage tracing","VAE-based watermarking","U-Net","latent vector transformation","model security","federated LDMs"],"ai_summary_model":"Qwen/Qwen2.5-Coder-32B-Instruct"},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"646c77911ee398a4e9404b8b","avatarUrl":"/avatars/05d1ea421dd4f3e2fd47cbe99fc52933.svg","isPro":false,"fullname":"Yunqiu Xu","user":"Yunqiu","type":"user"},{"_id":"62f39d617017a2dc33b28bbc","avatarUrl":"/avatars/2e325aa426c2a3d683eb30f098dd917b.svg","isPro":false,"fullname":"Yuan Gan","user":"thinice","type":"user"},{"_id":"695f3859b7deb674f80ca5aa","avatarUrl":"/avatars/560848ae8d19f0a40fa529677f1390a7.svg","isPro":false,"fullname":"GY","user":"NonameNowhere","type":"user"},{"_id":"6949066c8dee18ee17742339","avatarUrl":"/avatars/9343dc4c592a6dcfd5ce074de23f4333.svg","isPro":false,"fullname":"cheng","user":"wenlong831","type":"user"},{"_id":"6640c647acae6bb179eedff5","avatarUrl":"/avatars/bcaafaaa1d4b4c241d72a886401772e3.svg","isPro":false,"fullname":"Yuetong Liu","user":"YuetongLiu","type":"user"},{"_id":"695de72f0af94998f9fc4f91","avatarUrl":"/avatars/7beab1a3091a2d3bc201bff8b594f4fd.svg","isPro":false,"fullname":"Wu","user":"WZLWLM","type":"user"},{"_id":"6724e8dff0e98abd46c6319f","avatarUrl":"/avatars/9ac7018cd00dd44490e079b511413d02.svg","isPro":false,"fullname":"Yunze Wang","user":"unimodular","type":"user"},{"_id":"6523affbd89bc7773df3a765","avatarUrl":"/avatars/d7a2a21a93a51117c7372580d08383ae.svg","isPro":false,"fullname":"Yuchen Xian","user":"JINX-XIAN-528","type":"user"},{"_id":"65e59ecc63fb03ea72ef974b","avatarUrl":"/avatars/fad6bb396a9c35d3091f350842dfd418.svg","isPro":false,"fullname":"Liulei Li","user":"LLL-UTS","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2606/2606.22875.md","query":{}}">

Get this paper in your agent:

Don't have the latest CLI?

Cite arxiv.org/abs/2606.22875 in a model README.md to link it from this page.

Cite arxiv.org/abs/2606.22875 in a dataset README.md to link it from this page.

Cite arxiv.org/abs/2606.22875 in a Space README.md to link it from this page.

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