{"total":17,"items":[{"citing_arxiv_id":"2606.13140","ref_index":28,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"MIDSim: Simulating Multi-Channel Information Diffusion in Social Media with LLM-Powered Multi-Agent System","primary_cat":"cs.SI","submitted_at":"2026-06-11T10:05:11+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"MIDSim uses personalized LLM agents to jointly simulate social and algorithmic information diffusion streams and outperforms baselines on real diffusion events from three platforms.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.18264","ref_index":17,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Simulating Hate Speech Cascades with Multi-LLM Agents: Empirical Grounding, Modeling Fidelity, and Intervention Strategies","primary_cat":"cs.SI","submitted_at":"2026-05-21T21:58:26+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Multi-LLM agent simulations reproduce stance monoculture and toxicity-delta patterns from real Bluesky hateful cascades, with agent heterogeneity as the main fidelity driver and targeted interventions cutting spread 7.5-12.9% at 5.7% benign cost.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.18890","ref_index":67,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Stop Drawing Scientific Claims from LLM Social Simulations Without Robustness Audits","primary_cat":"physics.soc-ph","submitted_at":"2026-05-17T00:21:53+00:00","verdict":"ACCEPT","verdict_confidence":"MODERATE","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Minor perturbations in persona format, instruction framing, and network structure shift cooperation by up to 76 percentage points and polarization metrics consistently, showing that LLM social simulations require per-claim robustness audits via the new TRAILS taxonomy.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.12898","ref_index":5,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"When Do LLMs Generate Realistic Social Networks? A Multi-Dimensional Study of Culture, Language, Scale, and Method","primary_cat":"cs.SI","submitted_at":"2026-05-13T02:22:19+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"LLM social networks vary with culture, language, scale, and prompting method, matching real graphs on clustering but exceeding empirical demographic biases.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.12452","ref_index":30,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"The Algorithmic Caricature: Auditing LLM-Generated Political Discourse Across Crisis Events","primary_cat":"cs.CL","submitted_at":"2026-05-12T17:42:03+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"LLM-generated political discourse across crises is fluent yet caricatured: more negative, less emotionally varied, more structurally regular, and lexically abstract than observed online populations.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.10721","ref_index":12,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Conformity Generates Collective Misalignment in AI Agents Societies","primary_cat":"physics.soc-ph","submitted_at":"2026-05-11T15:30:48+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Populations of individually aligned AI agents reach stable misaligned states through conformity, with small adversarial agents able to trigger irreversible tipping points.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"interact in dedicated social networks, such as Chirper.ai or Moltbook.com, forming complex structures and statistical patterns typical of human social networks [14, 15]. Recent work has begun mapping the social and collective behavior of AI agents, revealing unexpected parallels to both hu- man psychology and physical systems [16-18]. Language models exhibit opinion dynamics [12, 19-22], build con- ventions through local interactions [23, 24], form complex networks [25, 26] and show the ability to spontaneously co- ordinate in large groups [27]. Only recently has attention ∗ giordano.de-marzo@uni-konstanz.de turned to the group-level properties of LLM populations as a safety concern in their own right [28]: collective interac-"},{"citing_arxiv_id":"2605.06196","ref_index":15,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"The Granularity Axis: A Micro-to-Macro Latent Direction for Social Roles in Language Models","primary_cat":"cs.AI","submitted_at":"2026-05-07T13:08:19+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"LLMs organize prompted social roles along a dominant, stable, and causally steerable granularity axis in representation space that runs from micro to macro levels.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"Recent large language models (LLMs) have demonstrated strong instruction following, open-ended interaction, and behavioral adaptation under prompting [ 1-5]. These capabilities have motivated growing interest in using LLMs to simulate human behavior and social interaction [6-8], including multi-agent environments [ 9-12] and domains such as politics [ 13, 14], public health [ 15], and markets [16, 17]. Compared with classical agent-based modeling, LLM-based simulation can elicit diverse behavioral patterns directly through language, but recent work also raises concerns about representational validity [18, 19], survey-response bias, and overly rationalized models of human decision-making [20, 21]. These concerns ultimately rest on what an LLM internally represents when"},{"citing_arxiv_id":"2604.25614","ref_index":77,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"HotComment: A Benchmark for Evaluating Popularity of Online Comments","primary_cat":"cs.AI","submitted_at":"2026-04-28T13:23:29+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"HotComment is a new multimodal benchmark that quantifies online comment popularity via content quality assessment, interaction-based prediction, and agent-simulated user engagement, accompanied by the StyleCmt stylistic model.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"arXiv:2412.15115 [cs.CL] https://arxiv.org/abs/2412.15115 [76] Fuchun Yang, Peng Li, Xin Guo, Yujie Zhang, and Shiyue Chen. 2019. Read- Attend-Comment: A Reading Comprehension Based Approach for Comment Generation. InProceedings of the Annual Meeting of the Association for Computa- tional Linguistics. Association for Computational Linguistics, 2567-2577. [77] Fuchun Yang, Peng Li, Qiang Wang, Yuan Xu, and Zhenhua Wei. 2019. Cross- Modal Comment Generation with Image and Text Features. InProceedings of the Annual Meeting of the Association for Computational Linguistics. Association for Computational Linguistics, 3456-3465. [78] Qianyun Yang, Zhiwei Chen, Yupeng Hu, Zixu Li, Zhiheng Fu, and Liqiang Nie."},{"citing_arxiv_id":"2604.11312","ref_index":57,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Network Effects and Agreement Drift in LLM Debates","primary_cat":"cs.SI","submitted_at":"2026-04-13T11:16:58+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"LLM agents in controlled network debates show agreement drift toward specific opinion positions, requiring separation of structural effects from LLM biases before using them as human behavioral proxies.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"with traditional agent-based modeling approaches [55], as demonstrated in [56], has shown how recommender systems can influence the quality of discourse. Specifically, agents exposed to popular content tend to exhibit increased toxicity and interparty engagement, mirroring the toxicity levels seen in U.S. Twitter discourse circa 2019. Further simulations reveal that confirmation bias among agents contributes to social fragmen- tation [57], aligning with established findings in opinion dynamics. LLMs are also capable of emulating persuasive communication [29, 58], generating coherent arguments based on psycho-linguistic theories of opinion change, and can be utilized for comprehensive social media simulations [18]. Nonetheless, LLMs have a natural inclination toward factual correctness and often resist generating content that contradicts"},{"citing_arxiv_id":"2604.03898","ref_index":10,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"LLM-Agent-based Social Simulation for Attitude Diffusion","primary_cat":"cs.AI","submitted_at":"2026-04-04T23:56:06+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"discourse_simulator is an open-source LLM-augmented agent-based modeling framework for simulating attitude diffusion on social networks in response to real-world events such as the 2025 Dublin anti-immigration march.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.20192","ref_index":44,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Leveraging Large Language Models for Sentiment Analysis: Multi-Modal Analysis of Decentraland's MANA Token","primary_cat":"cs.CL","submitted_at":"2026-04-04T04:04:13+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":3.0,"formal_verification":"none","one_line_summary":"Incorporating BERT-derived Discord sentiment into an LSTM improves MANA token return forecasts over a historical-price baseline.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2509.18985","ref_index":32,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Simulating Online Social Media Conversations on Controversial Topics Using AI Agents Calibrated on Real-World Data","primary_cat":"cs.SI","submitted_at":"2025-09-23T13:36:48+00:00","verdict":"CONDITIONAL","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"LLM agents calibrated on Italian election data produce coherent posts and realistic network structure but show less tone and toxicity variation than real users, with opinion changes resembling traditional mathematical models.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2509.06337","ref_index":35,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Large Language Models as Virtual Survey Respondents: Evaluating Sociodemographic Response Generation","primary_cat":"cs.AI","submitted_at":"2025-09-08T04:59:00+00:00","verdict":"CONDITIONAL","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Introduces PAS and FAS task abstractions plus the LLM-S^3 benchmark to evaluate LLMs on generating sociodemographic survey responses across 11 real datasets and multiple models.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2507.11521","ref_index":271,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Opinion dynamics: Statistical physics and beyond","primary_cat":"physics.soc-ph","submitted_at":"2025-07-15T17:45:12+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":2.0,"formal_verification":"none","one_line_summary":"A review synthesizing opinion dynamics research, categorizing models by macroscopic outcomes and microscopic mechanisms while connecting to empirical data and emerging AI tools.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2408.01257","ref_index":122,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Detection and Characterization of Coordinated Online Behavior: A Survey","primary_cat":"cs.SI","submitted_at":"2024-08-02T13:27:56+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"A literature survey that reconciles definitions of coordinated online behavior, proposes a study framework, reviews detection methods, and identifies research challenges.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"Progress in generative AI could thus make it harder to detect future instances of coordinated behavior or to assess the authenticity of coordinated actors. However, it is still unknown what the effect of these techniques will be on the landscape of online coordination. At the same time, generative AI (e.g., LLMs) can be used to simulate human behaviors within agent-based models [122], enabling the creation of realistic simulations of coordinated activities. In future, these models could be leveraged to train or evaluate detection methods, providing a controlled environment to probe methods' capabilities at detecting different instances and types of coordinated behavior. Manuscript submitted to ACM Detection and Characterization of Coordinated Online Behavior: A Survey 29"},{"citing_arxiv_id":"2402.05070","ref_index":123,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"A Roadmap to Pluralistic Alignment","primary_cat":"cs.AI","submitted_at":"2024-02-07T18:21:17+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"The paper formalizes three types of pluralistic AI models and three benchmark classes, arguing that current alignment techniques may reduce rather than increase distributional pluralism.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2305.09620","ref_index":95,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"AI-Augmented Surveys: Leveraging Large Language Models and Surveys for Opinion Prediction","primary_cat":"cs.CL","submitted_at":"2023-05-16T17:13:07+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"LLM embeddings enable strong retrodiction of masked GSS opinions via cross-validation and external validation but only modest performance on entirely unasked opinions.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null}],"limit":50,"offset":0}