Librarian Bot. I found the following papers similar to this paper.

The following papers were recommended by the Semantic Scholar API

iMAD: Intelligent Multi-Agent Debate for Efficient and Accurate LLM Inference (2025)
MUSIC: MUlti-Step Instruction Contrast for Multi-Turn Reward Models (2025)
Towards Proactive Personalization through Profile Customization for Individual Users in Dialogues (2025)
Incoherent Beliefs&Inconsistent Actions in Large Language Models (2025)
SpeakRL: Synergizing Reasoning, Speaking, and Acting in Language Models with Reinforcement Learning (2025)
PersonaMem-v2: Towards Personalized Intelligence via Learning Implicit User Personas and Agentic Memory (2025)
Adaptive Multi-Agent Response Refinement in Conversational Systems (2025)

Please give a thumbs up to this comment if you found it helpful!

If you want recommendations for any Paper on Hugging Face checkout this Space

You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: \n\n@librarian-bot\n\t recommend

\n","updatedAt":"2026-01-07T01:36:54.442Z","author":{"_id":"63d3e0e8ff1384ce6c5dd17d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg","fullname":"Librarian Bot (Bot)","name":"librarian-bot","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":318,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.7315990924835205},"editors":["librarian-bot"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg"],"reactions":[],"isReport":false}},{"id":"696be5ab776f8ff4db792f65","author":{"_id":"65243980050781c16f234f1f","avatarUrl":"/avatars/743a009681d5d554c27e04300db9f267.svg","fullname":"Avi","name":"avahal","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":3,"isUserFollowing":false},"createdAt":"2026-01-17T19:40:27.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"arXivlens breakdown of this paper 👉 https://arxivlens.com/PaperView/Details/confidence-estimation-for-llms-in-multi-turn-interactions-678-8dca17a6\n\n- Executive Summary\n- Detailed Breakdown\n- Practical Applications","html":"

arXivlens breakdown of this paper 👉 https://arxivlens.com/PaperView/Details/confidence-estimation-for-llms-in-multi-turn-interactions-678-8dca17a6

Executive Summary
Detailed Breakdown
Practical Applications

\n","updatedAt":"2026-01-17T19:40:27.600Z","author":{"_id":"65243980050781c16f234f1f","avatarUrl":"/avatars/743a009681d5d554c27e04300db9f267.svg","fullname":"Avi","name":"avahal","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":3,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.7219069600105286},"editors":["avahal"],"editorAvatarUrls":["/avatars/743a009681d5d554c27e04300db9f267.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2601.02179","authors":[{"_id":"695d1b7ac03d6d81e4399d2b","name":"Caiqi Zhang","hidden":false},{"_id":"695d1b7ac03d6d81e4399d2c","name":"Ruihan Yang","hidden":false},{"_id":"695d1b7ac03d6d81e4399d2d","name":"Xiaochen Zhu","hidden":false},{"_id":"695d1b7ac03d6d81e4399d2e","name":"Chengzu Li","hidden":false},{"_id":"695d1b7ac03d6d81e4399d2f","name":"Tiancheng Hu","hidden":false},{"_id":"695d1b7ac03d6d81e4399d30","name":"Yijiang River Dong","hidden":false},{"_id":"695d1b7ac03d6d81e4399d31","name":"Deqing Yang","hidden":false},{"_id":"695d1b7ac03d6d81e4399d32","name":"Nigel Collier","hidden":false}],"publishedAt":"2026-01-05T14:58:04.000Z","submittedOnDailyAt":"2026-01-06T11:57:24.455Z","title":"Confidence Estimation for LLMs in Multi-turn Interactions","submittedOnDailyBy":{"_id":"63920dfac47e36ddeb8f1864","avatarUrl":"/avatars/c36cbf7b084d62368312e5c9292e4260.svg","isPro":false,"fullname":"Caiqi Zhang","user":"caiqizh","type":"user"},"summary":"While confidence estimation is a promising direction for mitigating hallucinations in Large Language Models (LLMs), current research dominantly focuses on single-turn settings. The dynamics of model confidence in multi-turn conversations, where context accumulates and ambiguity is progressively resolved, remain largely unexplored. Reliable confidence estimation in multi-turn settings is critical for many downstream applications, such as autonomous agents and human-in-the-loop systems. This work presents the first systematic study of confidence estimation in multi-turn interactions, establishing a formal evaluation framework grounded in two key desiderata: per-turn calibration and monotonicity of confidence as more information becomes available. To facilitate this, we introduce novel metrics, including a length-normalized Expected Calibration Error (InfoECE), and a new \"Hinter-Guesser\" paradigm for generating controlled evaluation datasets. Our experiments reveal that widely-used confidence techniques struggle with calibration and monotonicity in multi-turn dialogues. We propose P(Sufficient), a logit-based probe that achieves comparatively better performance, although the task remains far from solved. Our work provides a foundational methodology for developing more reliable and trustworthy conversational agents.","upvotes":17,"discussionId":"695d1b7ac03d6d81e4399d33","githubRepo":"https://github.com/caiqizh/multi-turn-conf","githubRepoAddedBy":"auto","ai_summary":"Multi-turn conversation confidence estimation lacks systematic evaluation frameworks, prompting the introduction of novel metrics and a \"Hinter-Guesser\" paradigm for controlled dataset generation to improve calibration and monotonicity.","ai_keywords":["confidence estimation","large language models","multi-turn conversations","calibration","monotonicity","Expected Calibration Error","logit-based probe"],"githubStars":4,"organization":{"_id":"679c9d2bb741486264125a9a","name":"uni-cambridge","fullname":"University of Cambridge","avatar":"https://cdn-uploads.huggingface.co/production/uploads/679c98caac2b47d6f4132f9b/5tGEF02r7vvH94c0pQ5_0.png"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"645b0c3ec35da9c7afd95421","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/645b0c3ec35da9c7afd95421/vYBrCDagHsXAo6J2p-uG0.jpeg","isPro":false,"fullname":"Yuling","user":"YerbaPage","type":"user"},{"_id":"685a558b50a4d7e3c996e2e2","avatarUrl":"/avatars/50694a6bdf83a3b06512b4adcbf3409a.svg","isPro":false,"fullname":"Li Yuexian","user":"SDFHSIDFdFEF","type":"user"},{"_id":"63920dfac47e36ddeb8f1864","avatarUrl":"/avatars/c36cbf7b084d62368312e5c9292e4260.svg","isPro":false,"fullname":"Caiqi Zhang","user":"caiqizh","type":"user"},{"_id":"645474d788c97c8796dcb3d8","avatarUrl":"/avatars/dfab5c337c2c3d9c44149738f2e09517.svg","isPro":false,"fullname":"Beiduo CHen","user":"McmanusChen","type":"user"},{"_id":"671a4abbef737c0abe21b3f8","avatarUrl":"/avatars/da826af5472a3b9f1969f0c766672731.svg","isPro":false,"fullname":"Ruihan Yang","user":"rhyang2021","type":"user"},{"_id":"695d76bd94d1963e5dbf6e5c","avatarUrl":"/avatars/81998c86275b3db7b31fb8b40d08f96d.svg","isPro":false,"fullname":"James","user":"Jameshua","type":"user"},{"_id":"61b927af85a85a69ab914260","avatarUrl":"/avatars/5a8d2867c064423644101dbc74d863a5.svg","isPro":false,"fullname":"Tiancheng Hu","user":"pitehu","type":"user"},{"_id":"631efbfcc6b20f03c8211fc9","avatarUrl":"/avatars/3abf7acb7258c0f6d1a6016b293128b5.svg","isPro":false,"fullname":"txq","user":"future7","type":"user"},{"_id":"678f99215c1e705963aa8e26","avatarUrl":"/avatars/8be6e11af6ca818d090c8f420ae9c59d.svg","isPro":false,"fullname":"Yedidia AGNIMO","user":"YedsonUQ","type":"user"},{"_id":"689c92de47073c20aaa5664d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/3XFKakgZwLPmyGEi5MrJh.png","isPro":false,"fullname":"shebly","user":"yangzhichao","type":"user"},{"_id":"695c3d0dd0638f21b7f43f4e","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/NwnttXUqQmK0AwFyRxPRa.png","isPro":false,"fullname":"Kieran Garvey","user":"Atenai","type":"user"},{"_id":"620783f24e28382272337ba4","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/620783f24e28382272337ba4/zkUveQPNiDfYjgGhuFErj.jpeg","isPro":false,"fullname":"GuoLiangTang","user":"Tommy930","type":"user"}],"acceptLanguages":["*"],"dailyPaperRank":0,"organization":{"_id":"679c9d2bb741486264125a9a","name":"uni-cambridge","fullname":"University of Cambridge","avatar":"https://cdn-uploads.huggingface.co/production/uploads/679c98caac2b47d6f4132f9b/5tGEF02r7vvH94c0pQ5_0.png"}}">

Papers

arxiv:2601.02179

Confidence Estimation for LLMs in Multi-turn Interactions

Published on Jan 5

· Submitted by

Caiqi Zhang on Jan 6

University of Cambridge

Upvote

Authors:

Abstract

Multi-turn conversation confidence estimation lacks systematic evaluation frameworks, prompting the introduction of novel metrics and a "Hinter-Guesser" paradigm for controlled dataset generation to improve calibration and monotonicity.

AI-generated summary

While confidence estimation is a promising direction for mitigating hallucinations in Large Language Models (LLMs), current research dominantly focuses on single-turn settings. The dynamics of model confidence in multi-turn conversations, where context accumulates and ambiguity is progressively resolved, remain largely unexplored. Reliable confidence estimation in multi-turn settings is critical for many downstream applications, such as autonomous agents and human-in-the-loop systems. This work presents the first systematic study of confidence estimation in multi-turn interactions, establishing a formal evaluation framework grounded in two key desiderata: per-turn calibration and monotonicity of confidence as more information becomes available. To facilitate this, we introduce novel metrics, including a length-normalized Expected Calibration Error (InfoECE), and a new "Hinter-Guesser" paradigm for generating controlled evaluation datasets. Our experiments reveal that widely-used confidence techniques struggle with calibration and monotonicity in multi-turn dialogues. We propose P(Sufficient), a logit-based probe that achieves comparatively better performance, although the task remains far from solved. Our work provides a foundational methodology for developing more reliable and trustworthy conversational agents.