DebUnc: Improving Large Language Model Agent Communication With Uncertainty Metrics

TL;DR

DebUnc introduces a debate framework for large language models that incorporates uncertainty metrics to better communicate confidence, leading to improved accuracy in multi-agent discussions.

Contribution

It proposes a novel uncertainty-aware debate framework that enhances agent communication and accuracy by integrating confidence metrics into LLM interactions.

Findings

Attention-based methods outperform textual prompts in conveying confidence.

Performance improves with more reliable uncertainty estimation.

DebUnc enhances multi-agent debate accuracy.

Abstract

Multi-agent debates have been introduced to improve the accuracy of Large Language Models (LLMs) by having multiple agents discuss solutions to a problem over several rounds of debate. However, models often generate incorrect yet confident-sounding responses, which can mislead others. This issue arises partly because agents do not consider how confident their peers are. To address this, we propose DebUnc, a debate framework that uses uncertainty metrics to assess agent confidence. Confidence is then conveyed through a modified attention mechanism that adjusts token weights, or through textual prompts. Evaluations across benchmarks show that attention-based methods are particularly effective and that performance continues to improve as uncertainty estimation becomes more reliable. The code is available at https://github.com/lukeyoffe/debunc.