Most protein binder design methods assume a single target in a single structural state, while many real-world applications require one sequence to function across multiple conformations or targets. We introduce Chamaileon, a unified framework for cross-context binder design. Chamaileon combines In-Context Complex Co-Design (I3CD) for context-aware sequence-structure modeling with Mixture-of-Paths Sampling (MoPS), which iteratively optimizes a shared sequence across multiple structural contexts at inference time. We also introduce CROSS, a benchmark covering both multi-state and multi-target binder design. Our results show that Chamaileon can generate a single binder sequence that adapts its structure to satisfy distinct binding contexts, opening a path toward programmable multi-specific binders and conformational modulators. Code is available at <a href=\"https://github.com/caohengyuan/Chamaileon\" rel=\"nofollow\">https://github.com/caohengyuan/Chamaileon</a>.</p>\n","updatedAt":"2026-07-28T06:05:38.927Z","author":{"_id":"67e639a19310e4fdefebdf2c","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/snoVFf6SQX2SfwiEmDqqI.png","fullname":"caohengyuan","name":"caohy666","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":1,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8711044192314148},"editors":["caohy666"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/snoVFf6SQX2SfwiEmDqqI.png"],"reactions":[],"isReport":false}},{"id":"6a6846dc3168c93f6b7c0a01","author":{"_id":"6635b673e4028f5fd02b2131","avatarUrl":"/avatars/8c18cb14954fd2cd35f6975118471fba.svg","fullname":"YunhongLu","name":"JaydenLu666","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":4,"isUserFollowing":false},"createdAt":"2026-07-28T06:06:20.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"good paper!","html":"<p>good paper!</p>\n","updatedAt":"2026-07-28T06:06:20.490Z","author":{"_id":"6635b673e4028f5fd02b2131","avatarUrl":"/avatars/8c18cb14954fd2cd35f6975118471fba.svg","fullname":"YunhongLu","name":"JaydenLu666","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":4,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.6644483208656311},"editors":["JaydenLu666"],"editorAvatarUrls":["/avatars/8c18cb14954fd2cd35f6975118471fba.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2607.23518","authors":[{"_id":"6a68438a73f69d5af2bec935","user":{"_id":"67e639a19310e4fdefebdf2c","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/snoVFf6SQX2SfwiEmDqqI.png","isPro":false,"fullname":"caohengyuan","user":"caohy666","type":"user","name":"caohy666"},"name":"Hengyuan Cao","status":"claimed_verified","statusLastChangedAt":"2026-07-28T08:45:04.643Z","hidden":false},{"_id":"6a68438a73f69d5af2bec936","name":"Shizhuo Cheng","hidden":false},{"_id":"6a68438a73f69d5af2bec937","name":"Mingxuan Liu","hidden":false},{"_id":"6a68438a73f69d5af2bec938","name":"Weicheng Huang","hidden":false},{"_id":"6a68438a73f69d5af2bec939","name":"Yunhong Lu","hidden":false},{"_id":"6a68438a73f69d5af2bec93a","name":"Chenxi Cai","hidden":false},{"_id":"6a68438a73f69d5af2bec93b","name":"Yan Zhang","hidden":false},{"_id":"6a68438a73f69d5af2bec93c","name":"Min Zhang","hidden":false}],"publishedAt":"2026-07-26T00:00:00.000Z","submittedOnDailyAt":"2026-07-28T00:00:00.000Z","title":"Chamaileon: Cross-Context Binder Design with Contextualized Modeling and Mixed Sampling","submittedOnDailyBy":{"_id":"67e639a19310e4fdefebdf2c","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/snoVFf6SQX2SfwiEmDqqI.png","isPro":false,"fullname":"caohengyuan","user":"caohy666","type":"user","name":"caohy666"},"summary":"The rapid evolution of generative models has unlocked new potentials in protein binder design, a pivotal task in structural biology, by facilitating end-to-end generation via joint sequence-structure modeling or hallucination. However, existing approaches are predominantly implemented under a single-target, single-state assumption, limiting their ability to model multi-target or multi-state interactions required for advanced function-oriented protein design. Here, we introduce Chamaileon, which unifies multi-target and multi-state binder design by formulating the problem as cross-context binding landscape modeling. The framework is underpinned by a training paradigm termed In-Context Complex Co-Design (I3CD) for context-aware sequence-structure co-modeling. During inference, we employ Mixture-of-Paths Sampling (MoPS), a scalable strategy that optimizes a single sequence across contexts while alleviating the scarcity of high-quality multi-conformational paired data. Extensive evaluation on our newly constructed benchmark, CROSS, demonstrates that Chamaileon effectively generates sequences adaptable to diverse conformational landscapes and multi-target requirements. The code is available on https://github.com/caohengyuan/Chamaileon.","upvotes":6,"discussionId":"6a68438a73f69d5af2bec93d","githubRepo":"https://github.com/caohengyuan/Chamaileon","githubRepoAddedBy":"user","githubStars":12,"organization":{"_id":"6345aadf5efccdc07f1365a5","name":"ZhejiangUniversity","fullname":"Zhejiang University","avatar":"https://www.gravatar.com/avatar/d1d414628877bec2958f95ad283c15e7?d=retro&size=100"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"6635b673e4028f5fd02b2131","avatarUrl":"/avatars/8c18cb14954fd2cd35f6975118471fba.svg","isPro":false,"fullname":"YunhongLu","user":"JaydenLu666","type":"user"},{"_id":"67e639a19310e4fdefebdf2c","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/snoVFf6SQX2SfwiEmDqqI.png","isPro":false,"fullname":"caohengyuan","user":"caohy666","type":"user"},{"_id":"63465e923588a193ca4521ec","avatarUrl":"/avatars/9d856d87df825b91b18d0e33e336fe10.svg","isPro":false,"fullname":"Eternal Cheng","user":"Eternal2077","type":"user"},{"_id":"6599059f89010f9c7a26638b","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/N7BTaNAta7wM17WMKl-ZL.png","isPro":false,"fullname":"Jiwang Qu","user":"Tippye","type":"user"},{"_id":"663a6e68f1e6997947e28078","avatarUrl":"/avatars/f08ca525bbabb7781f77529235a0cd1a.svg","isPro":false,"fullname":"11","user":"DAI00111","type":"user"},{"_id":"69fc32753bc329c0e777ff5b","avatarUrl":"/avatars/9da44bacf7f84e4bde27d2f889800830.svg","isPro":false,"fullname":"OmniEdit-Bench","user":"OmniEdit-Bench","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"organization":{"_id":"6345aadf5efccdc07f1365a5","name":"ZhejiangUniversity","fullname":"Zhejiang University","avatar":"https://www.gravatar.com/avatar/d1d414628877bec2958f95ad283c15e7?d=retro&size=100"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2607/2607.23518.md","query":{}}">

Get this paper in your agent:

Don't have the latest CLI?

Cite arxiv.org/abs/2607.23518 in a model README.md to link it from this page.

Cite arxiv.org/abs/2607.23518 in a dataset README.md to link it from this page.

Cite arxiv.org/abs/2607.23518 in a Space README.md to link it from this page.

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