Debate or Vote: Multi-Agent LLM Decision Making
What it is
This is the official code repository for a NeurIPS 2025 Spotlight paper comparing debate and voting mechanisms in multi-agent LLM systems. It provides scripts to run experiments on various datasets using different LLM configurations, including sparse and centralized multi-agent debate setups.
The problem it solves
Lack of a clear, reproducible comparison between debate and voting mechanisms for improving decision quality in multi-agent LLM systems.
What you could use it for
- Reproducing the results of the 'Debate or Vote' paper
- Benchmarking multi-agent LLM decision-making strategies
- Testing heterogeneous agent configurations with multiple personas
Read the source ↗Back to the library →
Summary written by an AI model from the source; check the source before you rely on it.