Digital Pantheon: Simulating and Auditing Coalition Formation with LLM Agents
Dylan Van Mulders, Matthias Bogaert, Dirk Van den Poel
Why It Matters
What makes this one worth your time
Understanding coalition dynamics through simulation can aid political scientists and policymakers in exploring party compatibility and negotiation strategies before real-world applications.
A framework for simulating political coalition formation using LLM agents with ideological alignment.
Summary
The paper presents a multi-agent framework using LLMs to simulate and audit political coalition formation, focusing on the 2019 Flemish election. It combines Supervised Fine-Tuning, Direct Preference Optimization, and Retrieval-Augmented Generation to create partisan agents that negotiate coalition agreements. The framework includes a Multi-Layered Information Lineage Topology for tracing agreement clauses, a Coalition Influence Score for party impact assessment, and a real-world grounding pass for benchmarking against historical agreements.
Key contributions
- Development of a multi-agent framework for simulating political coalition formation.
- Integration of Supervised Fine-Tuning, Direct Preference Optimization, and Retrieval-Augmented Generation for ideological alignment.
- Introduction of a Multi-Layered Information Lineage Topology and Coalition Influence Score for agreement analysis.
Notable insights
- The use of Direct Preference Optimization to instill aggressive party-specific personas in LLM agents.
- The introduction of a Multi-Layered Information Lineage Topology to trace agreement clauses back to their manifesto origins.
Possible limitations
- Not stated in the abstract
Abstract
arXiv:2607.15095v2 Announce Type: replace-cross Abstract: The formation of political coalitions is a complex negotiation driven by both concrete policy objectives and deep-seated ideological convictions. While Large Language Models (LLMs) open new avenues for computational political science, the neutrality and helpfulness biases instilled by Reinforcement Learning from Human Feedback (RLHF) prevent them from sustaining steadfast partisan behaviour. We present a multi-agent framework that reconciles factual grounding with ideological alignment by combining Supervised Fine-Tuning (SFT), Direct Preference Optimization (DPO), and Retrieval-Augmented Generation (RAG): DPO instils aggressive party-specific personas, while a per-party RAG pipeline keeps each agent bounded to its official manifesto. We operationalize the framework on the 2019 Flemish election, deploying the partisan agents in a hub-and-spoke negotiation arbitrated by a formateur. To make the emergent negotiation interpretable, we introduce a Multi-Layered Information Lineage Topology (MILT) that traces every clause in the final agreement back to its manifesto origin and classifies it into five provenance states, a Coalition Influence Score (CIS) that aggregates these traceable contributions to identify which party shaped the agreement, and a real-world grounding pass that benchmarks each simulated provision against the historically adopted coalition agreement. Across three independent simulations the framework yields a stable winner and ranking (N-VA ahead of CD\&V and Open Vld), and manifesto-anchored lineage reliably predicts real-world materialization whereas hallucinated content does not. The result is a transparent, scalable testbed for the ex-ante exploration of party compatibility and formateur-mediated compromise.