Agent Sociodrama: Social-Dynamics-Guided Persona Recasting for LLM Agent Teams

Hugging Face Daily Papers Papers
llm-agents multi-agent-systems social-dynamics persona-engineering collaboration-benchmarks trajectory-analysis

Summary

Agent Sociodrama 是一个用于建模和验证 LLM 智能体团队中动态社会关系的框架，并提出 Trajectory-Guided Persona Recasting (TPR)，通过诊断交互轨迹中的群体失败来重塑智能体人设，在四个协作基准上最高提升 83.3% 的准确率，并在分布式信息合作与混合动机谈判任务上超越 SOTOPIA 和 AgentVerse。

Large language model (LLM) agents are increasingly deployed as teams, yet they fail in characteristically social ways: information holders stay silent, dominant speakers anchor groups on wrong answers, and norm violations go unsanctioned. We present Agent Sociodrama, a framework for modeling and validating evolving social dynamics in LLM agent teams. Under fixed personas, agents' emotional states and directed, asymmetric relationships are updated every round by a unified external LLM judge and shape their subsequent interactions. Across six LLM backbones and five social scenarios, we assess social fidelity through classical social-theory patterns and agreement with human judgments. The resulting trajectories exhibit structured social patterns and model-specific differences. We then introduce Trajectory-Guided Persona Recasting (TPR), which diagnoses group-dynamic failures from interaction transcripts and social trajectories and recasts agent personas before rerunning the task. Across four collaborative benchmarks, TPR improves performance from both heterogeneous and randomly initialized agent persona compositions, yielding up to 83.3%p of accuracy improvement over the initial casting. On representative distributed-information cooperation and mixed-motive bargaining tasks, we outperform SOTOPIA and AgentVerse in three of four comparisons, by up to 37.5%p in Pareto rate. These results show that validated social trajectories provide actionable signals for repairing LLM agent teams.
Original Article
View Cached Full Text

Cached at: 10/14/26, 06:54 AM

Paper page - Agent Sociodrama: Social-Dynamics-Guided Persona Recasting for LLM Agent Teams

Source: https://huggingface.co/papers/2610.15788

Abstract

Largelanguagemodel(LLM)agentsareincreasinglydeployedasteams,yettheyfailincharacteristicallysocialways:informationholdersstaysilent,dominantspeakersanchorgroupsonwronganswers,andnormviolationsgounsanctioned.WepresentAgentSociodrama,aframeworkformodelingandvalidatingevolvingsocialdynamicsinLLMagentteams.Underfixedpersonas,agents’emotionalstatesanddirected,asymmetricrelationshipsareupdatedeveryroundbyaunifiedexternalLLMjudgeandshapetheirsubsequentinteractions.AcrosssixLLMbackbonesandfivesocialscenarios,weassesssocialfidelitythroughclassicalsocial-theorypatternsandagreementwithhumanjudgments.Theresultingtrajectoriesexhibitstructuredsocialpatternsandmodel-specificdifferences.WethenintroduceTrajectory-GuidedPersonaRecasting(TPR),whichdiagnosesgroup-dynamicfailuresfrominteractiontranscriptsandsocialtrajectoriesandrecastsagentpersonasbeforererunningthetask.Acrossfourcollaborativebenchmarks,TPRimprovesperformancefrombothheterogeneousandrandomlyinitializedagentpersonacompositions,yieldingupto83.3%pofaccuracyimprovementovertheinitialcasting.Onrepresentativedistributed-informationcooperationandmixed-motivebargainingtasks,weoutperformSOTOPIAandAgentVerseinthreeoffourcomparisons,byupto37.5%pinParetorate.TheseresultsshowthatvalidatedsocialtrajectoriesprovideactionablesignalsforrepairingLLMagentteams.

View arXiv pageView PDFProject pageGitHub1Add to collection

Get this paper in your agent:

hf papers read 2610\.15788

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2610.15788 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2610.15788 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2610.15788 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

Persona-Guided LLM Agents for Task-Oriented Dialogue

arXiv cs.CL

This paper explores how LLMs can act as persona-guided agents in task-oriented dialogue, analyzing the effects of personality adaptation on task performance and user satisfaction through evaluations with models like GPT-4o and Gemini 2.0 Flash.

Point of Order: Action-Aware LLM Persona Modeling for Data-Grounded Civic Deliberation

arXiv cs.CL

A reproducible pipeline converts public Zoom recordings into speaker-attributed transcripts enriched with personas, topics, and action tags, then fine-tunes LLM personas on this data. Action-aware fine-tuning significantly improves persona fidelity, consistency, and deliberative responsiveness, enabling realistic civic deliberation simulations.