- \n
- Executive Summary \n
- Detailed Breakdown \n
- Practical Applications \n
This is an automated message from the Librarian Bot. I found the following papers similar to this paper.
\nThe following papers were recommended by the Semantic Scholar API
\n- \n
- Social Caption: Evaluating Social Understanding in Multimodal Models (2026) \n
- One Model, All Roles: Multi-Turn, Multi-Agent Self-Play Reinforcement Learning for Conversational Social Intelligence (2026) \n
- Small Symbols, Big Risks: Exploring Emoticon Semantic Confusion in Large Language Models (2026) \n
- M3-BENCH: Process-Aware Evaluation of LLM Agents Social Behaviors in Mixed-Motive Games (2026) \n
- HumanLLM: Towards Personalized Understanding and Simulation of Human Nature (2026) \n
- When Abundance Conceals Weakness: Knowledge Conflict in Multilingual Models (2026) \n
- S3IT: A Benchmark for Spatially Situated Social Intelligence Test (2025) \n
Please give a thumbs up to this comment if you found it helpful!
\nIf you want recommendations for any Paper on Hugging Face checkout this Space
\n You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: \n\n@librarian-bot\n\t recommend
arXiv explained breakdown of this paper 👉 https://arxivexplained.com/papers/socialveil-probing-social-intelligence-of-language-agents-under-communication-barriers
\n","updatedAt":"2026-02-08T18:03:35.578Z","author":{"_id":"65d9fc2a0e6ad24551d87a1e","avatarUrl":"/avatars/3aedb9522cc3cd08349d654f523fd792.svg","fullname":"Grant Singleton","name":"grantsing","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":4,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.6434029936790466},"editors":["grantsing"],"editorAvatarUrls":["/avatars/3aedb9522cc3cd08349d654f523fd792.svg"],"reactions":[],"isReport":false}},{"id":"698cceab5925dc396f4aa17e","author":{"_id":"664263ea8b2d38e53f04079c","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/664263ea8b2d38e53f04079c/_7og0ggbPpEHcH6g__DF0.jpeg","fullname":"Pengrui Han","name":"barryhpr","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":5,"isUserFollowing":false},"createdAt":"2026-02-11T18:47:07.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"this is a must read if you want to do social intelligence!!","html":"this is a must read if you want to do social intelligence!!
\n","updatedAt":"2026-02-11T18:47:07.226Z","author":{"_id":"664263ea8b2d38e53f04079c","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/664263ea8b2d38e53f04079c/_7og0ggbPpEHcH6g__DF0.jpeg","fullname":"Pengrui Han","name":"barryhpr","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":5,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.9529963135719299},"editors":["barryhpr"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/664263ea8b2d38e53f04079c/_7og0ggbPpEHcH6g__DF0.jpeg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2602.05115","authors":[{"_id":"698619f59c78be977c104bac","user":{"_id":"687ebb87416fcff56959d817","avatarUrl":"/avatars/8f13090dd6179bb18e8c8d205fd20131.svg","isPro":true,"fullname":"Keyang Xuan","user":"keyangx3","type":"user"},"name":"Keyang Xuan","status":"claimed_verified","statusLastChangedAt":"2026-02-09T08:34:55.860Z","hidden":false},{"_id":"698619f59c78be977c104bad","name":"Pengda Wang","hidden":false},{"_id":"698619f59c78be977c104bae","name":"Chongrui Ye","hidden":false},{"_id":"698619f59c78be977c104baf","name":"Haofei Yu","hidden":false},{"_id":"698619f59c78be977c104bb0","name":"Tal August","hidden":false},{"_id":"698619f59c78be977c104bb1","name":"Jiaxuan You","hidden":false}],"publishedAt":"2026-02-04T23:04:25.000Z","submittedOnDailyAt":"2026-02-06T14:46:49.653Z","title":"SocialVeil: Probing Social Intelligence of Language Agents under Communication Barriers","submittedOnDailyBy":{"_id":"687ebb87416fcff56959d817","avatarUrl":"/avatars/8f13090dd6179bb18e8c8d205fd20131.svg","isPro":true,"fullname":"Keyang Xuan","user":"keyangx3","type":"user"},"summary":"Large language models (LLMs) are increasingly evaluated in interactive environments to test their social intelligence. However, existing benchmarks often assume idealized communication between agents, limiting our ability to diagnose whether LLMs can maintain and repair interactions in more realistic, imperfect settings. To close this gap, we present SocialVeil, a social learning environment that can simulate social interaction under cognitive-difference-induced communication barriers. Grounded in a systematic literature review of communication challenges in human interaction, SocialVeil introduces three representative types of such disruption, semantic vagueness, sociocultural mismatch, and emotional interference. We also introduce two barrier-aware evaluation metrics, unresolved confusion and mutual understanding, to evaluate interaction quality under impaired communication. Experiments across 720 scenarios and four frontier LLMs show that barriers consistently impair performance, with mutual understanding reduced by over 45\\% on average, and confusion elevated by nearly 50\\%. Human evaluations validate the fidelity of these simulated barriers (ICCapprox0.78, Pearson rapprox0.80). We further demonstrate that adaptation strategies (Repair Instruction and Interactive learning) only have a modest effect far from barrier-free performance. This work takes a step toward bringing social interaction environments closer to real-world communication, opening opportunities for exploring the social intelligence of LLM agents.","upvotes":18,"discussionId":"698619f59c78be977c104bb2","ai_summary":"SocialVeil presents a social learning environment that simulates communication barriers in LLM interactions, demonstrating significant performance degradation under realistic conditions and limited effectiveness of adaptation strategies.","ai_keywords":["large language models","social intelligence","communication barriers","semantic vagueness","sociocultural mismatch","emotional interference","unresolved confusion","mutual understanding","interactive learning","repair instruction"]},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"679fff15d09622fbebbe1395","avatarUrl":"/avatars/109e6756cedb7d2b2d519564813db895.svg","isPro":false,"fullname":"Zhanyang Jin","user":"HolmesS","type":"user"},{"_id":"6705f27bbaae2d7ee2b4bf30","avatarUrl":"/avatars/ab79b73ca39ebc1793a2a8540d2ff1c6.svg","isPro":false,"fullname":"Zhang","user":"Diluner","type":"user"},{"_id":"664263ea8b2d38e53f04079c","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/664263ea8b2d38e53f04079c/_7og0ggbPpEHcH6g__DF0.jpeg","isPro":false,"fullname":"Pengrui Han","user":"barryhpr","type":"user"},{"_id":"65447a4ca3ab16953ba243eb","avatarUrl":"/avatars/50bd218c7e572d3572061a08ce793fcd.svg","isPro":false,"fullname":"Pengda Wang","user":"PengdaW","type":"user"},{"_id":"6675c9305eaa9dd299dcdca0","avatarUrl":"/avatars/504a259033605e489809c8f202538d75.svg","isPro":false,"fullname":"Zhouxiang Fang","user":"FocusV857","type":"user"},{"_id":"67b5693d3fe947554cfcacb7","avatarUrl":"/avatars/0f4527e6de5d614b38869a1cead11f52.svg","isPro":false,"fullname":"Cyrus","user":"uyffg","type":"user"},{"_id":"69861e91c9e8aa8e54c6779a","avatarUrl":"/avatars/ed5fd09bb7e921f5e160d8087b18a456.svg","isPro":false,"fullname":"Haotian Wu","user":"hw98","type":"user"},{"_id":"634b9914dcf125e4da02498b","avatarUrl":"/avatars/c3eff53d6dd0aa8990f6e54a51a4a510.svg","isPro":false,"fullname":"Chunyuan Deng","user":"CharlesDDDD","type":"user"},{"_id":"681b864112b9b68998430866","avatarUrl":"/avatars/47c068a526d1ecddbed9b2c08a00b430.svg","isPro":false,"fullname":"Lingxi Zhang","user":"zhanglingxi","type":"user"},{"_id":"649c5cf5c1ae48cf4d7dda34","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/649c5cf5c1ae48cf4d7dda34/bSJXATqkqBn8ypUy0OezY.jpeg","isPro":false,"fullname":"Peiyang Song","user":"p-song1","type":"user"},{"_id":"68f07d6b7dab5f6ec9ff891a","avatarUrl":"/avatars/df25ac1cd33edd7adc44758be1dfb079.svg","isPro":false,"fullname":"Fenghai Li","user":"4R5T","type":"user"},{"_id":"62c68fe4d9bb7ca028dc7323","avatarUrl":"/avatars/5114e6102c68da7ef367e63e61ce47dc.svg","isPro":false,"fullname":"Wenkai Li","user":"wenkai-li","type":"user"}],"acceptLanguages":["*"],"dailyPaperRank":0}">SocialVeil: Probing Social Intelligence of Language Agents under Communication Barriers
Abstract
SocialVeil presents a social learning environment that simulates communication barriers in LLM interactions, demonstrating significant performance degradation under realistic conditions and limited effectiveness of adaptation strategies.
Large language models (LLMs) are increasingly evaluated in interactive environments to test their social intelligence. However, existing benchmarks often assume idealized communication between agents, limiting our ability to diagnose whether LLMs can maintain and repair interactions in more realistic, imperfect settings. To close this gap, we present SocialVeil, a social learning environment that can simulate social interaction under cognitive-difference-induced communication barriers. Grounded in a systematic literature review of communication challenges in human interaction, SocialVeil introduces three representative types of such disruption, semantic vagueness, sociocultural mismatch, and emotional interference. We also introduce two barrier-aware evaluation metrics, unresolved confusion and mutual understanding, to evaluate interaction quality under impaired communication. Experiments across 720 scenarios and four frontier LLMs show that barriers consistently impair performance, with mutual understanding reduced by over 45\% on average, and confusion elevated by nearly 50\%. Human evaluations validate the fidelity of these simulated barriers (ICCapprox0.78, Pearson rapprox0.80). We further demonstrate that adaptation strategies (Repair Instruction and Interactive learning) only have a modest effect far from barrier-free performance. This work takes a step toward bringing social interaction environments closer to real-world communication, opening opportunities for exploring the social intelligence of LLM agents.
Community
Interesting work, Keyang
Very insightful Keyang! Great work!
We introduce SOCIALVEIL, an interactive social learning environment that can simulate social interaction under cognitive-difference-induced communication barriers.
Congrats on the great work Keyang!
Very exciting work, Keyang!
arXivLens breakdown of this paper 👉 https://arxivlens.com/PaperView/Details/socialveil-probing-social-intelligence-of-language-agents-under-communication-barriers-855-67ea3c30
- Executive Summary
- Detailed Breakdown
- Practical Applications
This is an automated message from the Librarian Bot. I found the following papers similar to this paper.
The following papers were recommended by the Semantic Scholar API
- Social Caption: Evaluating Social Understanding in Multimodal Models (2026)
- One Model, All Roles: Multi-Turn, Multi-Agent Self-Play Reinforcement Learning for Conversational Social Intelligence (2026)
- Small Symbols, Big Risks: Exploring Emoticon Semantic Confusion in Large Language Models (2026)
- M3-BENCH: Process-Aware Evaluation of LLM Agents Social Behaviors in Mixed-Motive Games (2026)
- HumanLLM: Towards Personalized Understanding and Simulation of Human Nature (2026)
- When Abundance Conceals Weakness: Knowledge Conflict in Multilingual Models (2026)
- S3IT: A Benchmark for Spatially Situated Social Intelligence Test (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:
@librarian-bot
recommend
arXiv explained breakdown of this paper 👉 https://arxivexplained.com/papers/socialveil-probing-social-intelligence-of-language-agents-under-communication-barriers
this is a must read if you want to do social intelligence!!
Models citing this paper 0
No model linking this paper
Datasets citing this paper 0
No dataset linking this paper
Spaces citing this paper 0
No Space linking this paper