Paper Accepted at EMNLP 2026

Our paper on simulating democratic deliberation with multi-agent large language models has been accepted at EMNLP 2026, the Conference on Empirical Methods in Natural Language Processing, to be held in Budapest on 24–29 October 2026. The work introduces VoteSim, a framework for representing pluralistic preferences through electoral systems in populations of LLM agents.

Simulating Democratic Deliberation: Representing Pluralistic Preferences through Electoral Systems in Multi-Agent LLMs
Faulkner R, Szufa S, Hoyer D, Zhang TJ, Bouffanais R, Leibo JZ & Jin Z
EMNLP 2026, Budapest, Hungary, October 2026. [pdf] [code]