Bayesian Belief Layer for Controllable Opinion Dynamics in LLM Agents
On the map: LLM agents and tool use → Social simulation and multi-agent debate
Abstract (arXiv)
LLM agents in social simulation revise their opinions implicitly, in context: how open an agent is to persuasion can neither be specified nor verified, and collective outcomes inherit the model's training prior. We introduce Bayesian Chronicle Agents (BCA), a minimal belief layer separating what an agent believes from how it speaks. Each stance is a probability, updated by one Bayesian step per utterance heard. A single prior-strength parameter kappa encodes stubbornness, modeled after its role in Friedkin--Johnsen (FJ) opinion dynamics. We then sweep this parameter to yield three canonical regimes of opinion dynamics on demand (consensus, persistent disagreement, committed-minority influence), with persistent disagreement matching the FJ closed-form fixed points at R2!=!0.93--0.99. We further show that prescribed kappa remains recoverable after the language round-trip, with perfect rank-order recovery across all four models. Explicit belief also makes simulation auditable: the layer surfaces systematic per-model stance biases that end-to-end simulation would silently absorb.
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Social simulation and multi-agent debate, one of 16 directions in LLM agents and tool use. Builds on 6 of its 14 references that are in the map. First posted on arXiv 18 Sep 2026.
Builds on
6 of 14 references.
Closest ancestors · 3
Same direction, newest first
- 2026Belief Engine: Configurable and Inspectable Stance Dynamics in Multi-Agent LLM Deliberation14 May
- 2025ID-RAG: Identity Retrieval-Augmented Generation for Long-Horizon Persona Coherence in Generative Agents29 Sep
- 2024Systematic Biases in LLM Simulations of Debates6 Feb
Other references · 3
- 2017On Calibration of Modern Neural Networkscited by 4,860
- 2024Emergent social conventions and collective bias in LLM populationscited by 115
- 2025From Evidence to Belief: A Bayesian Epistemology Approach to Language Modelscited by 5
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2026ScioMind: Cognitively Grounded Multi-Agent Social Simulation with Anchoring-Based Belief Dynamics and Dynamic Profiles
- 2023Simulating Opinion Dynamics with Networks of LLM-based Agents
- 2025DEBATE: A Large-Scale Benchmark for Evaluating Opinion Dynamics in Role-Playing LLM Agents
- 2025Disentangling Interaction and Bias Effects in Opinion Dynamics of Large Language Models
- 2026LLM Agents Make Collective Belief Dynamics Programmable: Challenges and Research Directions
- 2024On the Principles behind Opinion Dynamics in Multi-Agent Systems of Large Language Models
- 2025When Your AI Agent Succumbs to Peer-Pressure: Studying Opinion-Change Dynamics of LLMs
- 2026BeliefShift: Benchmarking Temporal Belief Consistency and Opinion Drift in LLM Agents