Autonomous negotiation agents are moving into high-stakes settings like insurance and procurement. Everyone's been protecting the values being exchanged with cryptography — but we show that's not enough: the negotiation behavior itself (how you concede, when, and how fast) leaks your private constraints, even when every number on the wire is encrypted. We formalize this as behavioral differential privacy for multi-round negotiation, and design an adaptive stochastic policy that jointly guarantees (ε, δ)-DP, almost-sure convergence to agreement (when a deal is possible), and high utility. Across 3,000 synthetic bilateral negotiations it cuts an adversary's inference accuracy by 43-50% while keeping success rate and utility above 90%.</p>\n<p> Takeaway: you can hide your negotiation strategy without losing the deal.</p>\n","updatedAt":"2026-07-20T16:06:34.031Z","author":{"_id":"6a529655066028dd8a66ec52","avatarUrl":"/avatars/a2d92ebb13bb102fe59309f4db2c3e36.svg","fullname":"Barkha Rani","name":"ranibarkha","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":1,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.9324339628219604},"editors":["ranibarkha"],"editorAvatarUrls":["/avatars/a2d92ebb13bb102fe59309f4db2c3e36.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2607.06815","authors":[{"_id":"6a529a9da9d74d6e65bbc916","user":{"_id":"6a529655066028dd8a66ec52","avatarUrl":"/avatars/a2d92ebb13bb102fe59309f4db2c3e36.svg","isPro":false,"fullname":"Barkha Rani","user":"ranibarkha","type":"user","name":"ranibarkha"},"name":"Barkha Rani","status":"claimed_verified","statusLastChangedAt":"2026-07-20T15:45:23.179Z","hidden":false}],"mediaUrls":["https://cdn-uploads.huggingface.co/production/uploads/6a529655066028dd8a66ec52/H9txWNxIkULmAfnHeHj6i.png"],"publishedAt":"2026-07-07T00:00:00.000Z","submittedOnDailyAt":"2026-07-20T00:00:00.000Z","title":"Behavioral Privacy Leakage in Agentic Negotiation: Formalizing and Mitigating Inference Attacks via Randomized Policies","submittedOnDailyBy":{"_id":"6a529655066028dd8a66ec52","avatarUrl":"/avatars/a2d92ebb13bb102fe59309f4db2c3e36.svg","isPro":false,"fullname":"Barkha Rani","user":"ranibarkha","type":"user","name":"ranibarkha"},"summary":"Autonomous negotiation agents are increasingly deployed in high-stakes settings such as insurance and procurement. While cryptographic techniques protect explicitly disclosed constraint values, they fail to address a subtler threat: behavioral privacy leakage, where an adversary infers private constraints from observable negotiation dynamics such as concession trajectories, timing, and convergence patterns. This paper investigates behavioral differential privacy in multi-round negotiation protocols. We design an adaptive stochastic negotiation policy that jointly guarantees (varepsilon, δ)-differential privacy, almost-sure convergence of the offer sequence (reaching agreement when the counterparty's reservation value permits), and high negotiation utility. Evaluated on 3,000 synthetic bilateral negotiations, our mechanism reduces adversarial inference accuracy by 43-50% while maintaining a negotiation success rate and utility above 90%, demonstrating that strong privacy guarantees can be achieved without significant loss of performance.","upvotes":4,"discussionId":"6a529a9da9d74d6e65bbc917"},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"6a529655066028dd8a66ec52","avatarUrl":"/avatars/a2d92ebb13bb102fe59309f4db2c3e36.svg","isPro":false,"fullname":"Barkha Rani","user":"ranibarkha","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":"6a2da6c8ca070ee12c6e396c","avatarUrl":"/avatars/0355287dcabaa67dbc7f0b10b87451f9.svg","isPro":false,"fullname":"Joe Mama","user":"JoeMama123123123","type":"user"},{"_id":"631e14ac473a6825f285e89d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/631e14ac473a6825f285e89d/K-6QnoeGLg8XFvbTMMdqA.jpeg","isPro":false,"fullname":"Yury Panikov","user":"panikov","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2607/2607.06815.md","query":{}}">
Behavioral Privacy Leakage in Agentic Negotiation: Formalizing and Mitigating Inference Attacks via Randomized Policies
Abstract
Autonomous negotiation agents are increasingly deployed in high-stakes settings such as insurance and procurement. While cryptographic techniques protect explicitly disclosed constraint values, they fail to address a subtler threat: behavioral privacy leakage, where an adversary infers private constraints from observable negotiation dynamics such as concession trajectories, timing, and convergence patterns. This paper investigates behavioral differential privacy in multi-round negotiation protocols. We design an adaptive stochastic negotiation policy that jointly guarantees (varepsilon, δ)-differential privacy, almost-sure convergence of the offer sequence (reaching agreement when the counterparty's reservation value permits), and high negotiation utility. Evaluated on 3,000 synthetic bilateral negotiations, our mechanism reduces adversarial inference accuracy by 43-50% while maintaining a negotiation success rate and utility above 90%, demonstrating that strong privacy guarantees can be achieved without significant loss of performance.
Community
Autonomous negotiation agents are moving into high-stakes settings like insurance and procurement. Everyone's been protecting the values being exchanged with cryptography — but we show that's not enough: the negotiation behavior itself (how you concede, when, and how fast) leaks your private constraints, even when every number on the wire is encrypted. We formalize this as behavioral differential privacy for multi-round negotiation, and design an adaptive stochastic policy that jointly guarantees (ε, δ)-DP, almost-sure convergence to agreement (when a deal is possible), and high utility. Across 3,000 synthetic bilateral negotiations it cuts an adversary's inference accuracy by 43-50% while keeping success rate and utility above 90%.
Takeaway: you can hide your negotiation strategy without losing the deal.
Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.
Tap or paste here to upload images
Cite arxiv.org/abs/2607.06815 in a model README.md to link it from this page.
Cite arxiv.org/abs/2607.06815 in a dataset README.md to link it from this page.
Cite arxiv.org/abs/2607.06815 in a Space README.md to link it from this page.
Discussion (0)
Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.
Sign in →No comments yet. Sign in and be the first to say something.