SODE: Analyzing Social Dynamics in LLM Agents

• URL: http://arxiv.org/abs/2605.23949v1
• Date: Wed, 06 May 2026 14:50:07 GMT
• Title: SODE: Analyzing Social Dynamics in LLM Agents
• Abstract summary: We introduce SODE (Social Dynamics Evaluation), a framework that evaluates AI agents across three evolutionary dimensions.<n>Applying SODE reveals systematic divergences.<n>We demonstrate that a "long-horizon framing" can unlock reciprocal capabilities in reasoning models.
• Score: 17.885513252130373
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: As Large Language Models (LLMs) evolve into interactive agents, understanding their behavioral alignment within human social dynamics becomes essential. While behavioral game theory offers a framework to study these interactions, previous work has predominantly relied on outcome-based metrics such as average scores. This focus overlooks the mechanisms that facilitate sustainable cooperation, as identical scores can be derived from vastly different strategies. To bridge this gap, we introduce SODE (Social Dynamics Evaluation), a framework that evaluates LLM agents across three evolutionary dimensions: Direct Reciprocity for strategy adaptation, Indirect Reciprocity for reputation sensitivity, and Group Dynamics for cooperative resilience. Applying SODE reveals systematic divergences: instruction-tuned models often exhibit "passive compliance" that renders them vulnerable to exploitation, while reasoning models prioritize short-horizon optimization, destabilizing long-term cooperation. Notably, we demonstrate that a "long-horizon framing" can unlock reciprocal capabilities in reasoning models. Thus, SODE offers a systematic, mechanism-grounded benchmark for aligning AI agents with complex human social dynamics.

Related papers

• MobEvolve: An Agentic Self-Evolving Heuristic System for Interpretable Human Mobility Generation [51.824145170051516]
  MobEvolve is a self-evolving framework for human mobility generation.<n>It employs an agent to iteratively evolve its internal logic.<n>It significantly outperforms state-of-the-art deep generative and LLM-based methods.
  arXiv  Detail & Related papers  (2026-06-01T03:46:25Z)
• ALSO: Adversarial Online Strategy Optimization for Social Agents [63.84788516416339]
  We propose textbfALSO (textbfAdrial ontextbfLine textbfStrategy textbfOptimization) as the first framework for online strategy optimization in social simulation.<n> ALSO formulates multi-turn interaction as an adversarial bandit problem, where combinations of static personas and dynamic strategy instructions are treated as arms.<n>Also introduces a lightweight neural surrogate to predict rewards from interaction histories, enabling sample-efficient exploration and continuous online adaptation.
  arXiv  Detail & Related papers  (2026-05-15T09:25:15Z)
• Understanding LLM Agent Behaviours via Game Theory: Strategy Recognition, Biases and Multi-Agent Dynamics [1.6487772637295166]
  We extend the FAIRGAME framework to evaluate Large Language Models (LLMs) behaviour in repeated social dilemmas.<n>We show that LLMs exhibit systematic, model- and language-dependent behavioural intentions, with linguistic framing at times exerting effects as strong as architectural differences.
  arXiv  Detail & Related papers  (2025-12-08T11:40:03Z)
• Social World Model-Augmented Mechanism Design Policy Learning [58.739456918502704]
  We introduce SWM-AP (Social World Model-Augmented Mechanism Design Policy Learning), which learns a social world model hierarchically to enhance mechanism design.<n>We show that SWM-AP outperforms established model-based and model-free RL baselines in cumulative rewards and sample efficiency.
  arXiv  Detail & Related papers  (2025-10-22T06:01:21Z)
• The Landscape of Agentic Reinforcement Learning for LLMs: A Survey [103.32591749156416]
  The emergence of agentic reinforcement learning (Agentic RL) marks a paradigm shift from conventional reinforcement learning applied to large language models (LLM RL)<n>This survey formalizes this conceptual shift by contrasting the degenerate single-step Markov Decision Processes (MDPs) of LLM-RL with the temporally extended, partially observable Markov decision processes (POMDPs) that define Agentic RL.
  arXiv  Detail & Related papers  (2025-09-02T17:46:26Z)
• Simulating Generative Social Agents via Theory-Informed Workflow Design [11.992123170134185]
  We propose a theory-informed framework that provides a systematic design process for social agents.<n>Our framework is grounded in principles from Social Cognition Theory and introduces three key modules: motivation, action planning, and learning.<n>Experiments demonstrate that our theory-driven agents reproduce realistic human behavior patterns under complex conditions.
  arXiv  Detail & Related papers  (2025-08-12T08:14:48Z)
• DynamiX: Large-Scale Dynamic Social Network Simulator [101.65679342680542]
  DynamiX is a novel large-scale social network simulator dedicated to dynamic social network modeling.<n>For opinion leaders, we propose an information-stream-based link prediction method recommending potential users with similar stances.<n>For ordinary users, we construct an inequality-oriented behavior decision-making module.
  arXiv  Detail & Related papers  (2025-07-26T12:13:30Z)
• GenAI-based Multi-Agent Reinforcement Learning towards Distributed Agent Intelligence: A Generative-RL Agent Perspective [35.589506360952925]
  We argue for a transformative paradigm shift from reactive to proactive multi-agent intelligence through generative AI-based reinforcement learning.<n>Rather than responding to immediate observations, generative-RL agents can model environment evolution, predict other agents' behaviors, generate coordinated action sequences, and engage in strategic reasoning accounting for long-term dynamics.
  arXiv  Detail & Related papers  (2025-07-13T05:02:43Z)
• A Grounded Observer Framework for Establishing Guardrails for Foundation Models in Socially Sensitive Domains [1.9116784879310025]
  Given the complexities of foundation models, traditional techniques for constraining agent behavior cannot be directly applied.<n>We propose a grounded observer framework for constraining foundation model behavior that offers both behavioral guarantees and real-time variability.
  arXiv  Detail & Related papers  (2024-12-23T22:57:05Z)
• Multi-Agent Dynamic Relational Reasoning for Social Robot Navigation [50.01551945190676]
  Social robot navigation can be helpful in various contexts of daily life but requires safe human-robot interactions and efficient trajectory planning. We propose a systematic relational reasoning approach with explicit inference of the underlying dynamically evolving relational structures. We demonstrate its effectiveness for multi-agent trajectory prediction and social robot navigation.
  arXiv  Detail & Related papers  (2024-01-22T18:58:22Z)
• An active inference model of collective intelligence [0.0]
  This paper posits a minimal agent-based model that simulates the relationship between local individual-level interaction and collective intelligence. Results show that stepwise cognitive transitions increase system performance by providing complementary mechanisms for alignment between agents' local and global optima.
  arXiv  Detail & Related papers  (2021-04-02T14:32:01Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.