# Belief Engine: Configurable and Inspectable Stance Dynamics in Multi-Agent LLM Deliberation

arXiv: https://arxiv.org/abs/2605.15343 · HTML: https://mlsanity.com/paper/2605.15343/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Social simulation and multi-agent debate, one of 16 directions in LLM agents and tool use.
- Builds on 24 of its 40 references that are in the map.
- First posted on arXiv 14 May 2026.

## Placed in

- LLM agents and tool use / [Social simulation and multi-agent debate](https://mlsanity.com/agents/social-simulation-and-multi-agent-debate/)

## Builds on (most cited first, 24 of 24)

- [Generative Agents: Interactive Simulacra of Human Behavior](https://mlsanity.com/paper/2304.03442/) (2023)
- [DeBERTaV3: Improving DeBERTa using ELECTRA-Style Pre-Training with Gradient-Disentangled Embedding Sharing](https://mlsanity.com/paper/2111.09543/) (2021)
- [A-MEM: Agentic Memory for LLM Agents](https://mlsanity.com/paper/2502.12110/) (2025)
- [Knowledge Conflicts for LLMs: A Survey](https://mlsanity.com/paper/2403.08319/) (2024)
- [HippoRAG: Neurobiologically Inspired Long-Term Memory for Large Language Models](https://mlsanity.com/paper/2405.14831/) (2024)
- [AgentSociety: Large-Scale Simulation of LLM-Driven Generative Agents Advances Understanding of Human Behaviors and Society](https://mlsanity.com/paper/2502.08691/) (2025)
- [Memory OS of AI Agent](https://mlsanity.com/paper/2506.06326/) (2025)
- [How Memory Management Impacts LLM Agents: An Empirical Study of Experience-Following Behavior](https://mlsanity.com/paper/2505.16067/) (2025)
- [Systematic Biases in LLM Simulations of Debates](https://mlsanity.com/paper/2402.04049/) (2024)
- [Debate or Vote: Which Yields Better Decisions in Multi-Agent Large Language Models?](https://mlsanity.com/paper/2508.17536/) (2025)
- [A Large-scale Dataset for Argument Quality Ranking: Construction and Analysis](https://mlsanity.com/paper/1911.11408/) (2019)
- [LLM Voting: Human Choices and AI Collective Decision Making](https://mlsanity.com/paper/2402.01766/) (2024)
- [LiCoMemory: Lightweight and Cognitive Agentic Memory for Efficient Long-Term Reasoning](https://mlsanity.com/paper/2511.01448/) (2025)
- [From Debate to Equilibrium: Belief-Driven Multi-Agent LLM Reasoning via Bayesian Nash Equilibrium](https://mlsanity.com/paper/2506.08292/) (2025)
- [Examining Identity Drift in Conversations of LLM Agents](https://mlsanity.com/paper/2412.00804/) (2024)
- [Multi-Agent Debate for LLM Judges with Adaptive Stability Detection](https://mlsanity.com/paper/2510.12697/) (2025)
- [Reasoning Models Generate Societies of Thought](https://mlsanity.com/paper/2601.10825/) (2026)
- [AI Debate Aids Assessment of Controversial Claims](https://mlsanity.com/paper/2506.02175/) (2025)
- [Echoing: Identity Failures when LLM Agents Talk to Each Other](https://mlsanity.com/paper/2511.09710/) (2025)
- [From Evidence to Belief: A Bayesian Epistemology Approach to Language Models](https://mlsanity.com/paper/2504.19622/) (2025)
- [R-Debater: Retrieval-Augmented Debate Generation through Argumentative Memory](https://mlsanity.com/paper/2512.24684/) (2025)
- [Large Language Models as Discounted Bayesian Filters](https://mlsanity.com/paper/2512.18489/) (2025)
- [Topology-Aware LLM-Driven Social Simulation: A Unified Framework for Efficient and Realistic Agent Dynamics](https://mlsanity.com/paper/2604.18011/) (2026)
- [Agora: Teaching the Skill of Consensus-Finding with AI Personas Grounded in Human Voice](https://mlsanity.com/paper/2603.07339/) (2026)

## Built on by (most cited first, 2 of 2)

- [When Does Belief-Based Agent Memory Help? Reliability-Conditional Updating and Provenance-Capped Poisoning Defense](https://mlsanity.com/paper/2606.22030/) (2026)
- [Bayesian Belief Layer for Controllable Opinion Dynamics in LLM Agents](https://mlsanity.com/paper/2609.21997/) (2026)

## Abstract (arXiv)

LLM-based agents are increasingly used to simulate deliberative interactions such as negotiation, conflict resolution, and multi-turn opinion exchange. Yet generated transcripts often do not reveal why an agent's stance changes: movement may reflect evidence uptake, anchoring, role drift, echoing, or changed prompt and retrieval context. We introduce the Belief Engine (BE), an auditable belief-update layer that treats "belief" as an evidential state over a proposition and exposes it as scalar stance. BE extracts arguments into structured memory and updates stance with a log-odds rule controlled by evidence uptake u and prior anchoring a. Across multiple base LLMs, parameter sweeps show that these controls reliably shape stance dynamics while preserving an evidence-level update trail. On DEBATE, a human deliberation dataset with pre/post opinions, BE best reconstructs participants whose final stance follows extracted evidence; stable and evidence-opposed cases instead point to anchoring or factors outside the extracted evidence stream. BE provides configurable infrastructure for studying evidence-grounded deliberation, where openness, commitment, convergence, and disagreement can be tied to explicit update assumptions rather than hidden prompt effects.