{"paper":{"title":"S$^3$: Social-network Simulation System with Large Language Model-Empowered Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"LLM agents in the S3 system emulate human perception and actions to produce emergent social network phenomena like information and emotion propagation.","cross_cats":[],"primary_cat":"cs.SI","authors_text":"Chen Gao, Depeng Jin, Huandong Wang, Jinghua Piao, Jinzhu Mao, Xiaochong Lan, Yong Li, Zhihong Lu","submitted_at":"2023-07-27T16:24:56Z","abstract_excerpt":"Social network simulation plays a crucial role in addressing various challenges within social science. It offers extensive applications such as state prediction, phenomena explanation, and policy-making support, among others. In this work, we harness the formidable human-like capabilities exhibited by large language models (LLMs) in sensing, reasoning, and behaving, and utilize these qualities to construct the S$^3$ system (short for $\\textbf{S}$ocial network $\\textbf{S}$imulation $\\textbf{S}$ystem). Adhering to the widely employed agent-based simulation paradigm, we employ prompt engineering "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"By endowing the agent in the system with the ability to perceive the informational environment and emulate human actions, we observe the emergence of population-level phenomena, including the propagation of information, attitudes, and emotions. ... the results demonstrate promising accuracy.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That prompt engineering and prompt tuning suffice to make LLM agents emulate genuine human behavior in social networks closely enough for the observed population-level phenomena to be meaningful.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"S³ uses LLM agents to simulate social networks by modeling emotion, attitude, and interaction, producing emergent propagation phenomena with promising accuracy on real data.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"LLM agents in the S3 system emulate human perception and actions to produce emergent social network phenomena like information and emotion propagation.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"41f9cb1a21840b1c49e35a548272e18529598ec3c134eb2266a48852ed5a08e7"},"source":{"id":"2307.14984","kind":"arxiv","version":3},"verdict":{"id":"01ab8e6f-bf72-42f7-b90c-dbff6365182e","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-17T11:24:12.286746Z","strongest_claim":"By endowing the agent in the system with the ability to perceive the informational environment and emulate human actions, we observe the emergence of population-level phenomena, including the propagation of information, attitudes, and emotions. ... the results demonstrate promising accuracy.","one_line_summary":"S³ uses LLM agents to simulate social networks by modeling emotion, attitude, and interaction, producing emergent propagation phenomena with promising accuracy on real data.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That prompt engineering and prompt tuning suffice to make LLM agents emulate genuine human behavior in social networks closely enough for the observed population-level phenomena to be meaningful.","pith_extraction_headline":"LLM agents in the S3 system emulate human perception and actions to produce emergent social network phenomena like information and emotion propagation."},"references":{"count":46,"sample":[{"doi":"","year":2023,"title":"Using large language models to simulate multiple humans and replicate human subject studies","work_id":"ef38a5c8-c332-4eaf-837a-e4df61e80bfb","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1997,"title":"Advancing the art of simulation in the social sciences","work_id":"3f5bbbae-1125-4776-a895-72cb6e4d1e5d","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2020,"title":"Modeling echo chambers and polarization dynamics in social networks.Physical Review Letters, 124(4):048301, 2020","work_id":"a41d7aa4-bf3c-4f0f-86ea-5cb97ba074de","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2021,"title":"Emer- gence of polarized ideological opinions in multidimensional topic spaces","work_id":"3b5d2dcf-b995-499a-8151-9b094e965707","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1987,"title":"A guide to simulation, 1987","work_id":"38c44066-99fc-40a0-9adf-5e69ac837882","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":46,"snapshot_sha256":"c52732ed388bcbb439d5a7115947118ae206b8be45ae3355c8e11f39f2175cb5","internal_anchors":4},"formal_canon":{"evidence_count":2,"snapshot_sha256":"e2bc64f1f25231ffaf75ecb8daae37ba3b4722b576e1bcac19cf06be2a04be9f"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}