DebUnc: Improving Large Language Model Agent Communication With Uncertainty Metrics

2024-07-08 · Luke Yoffe, Alfonso Amayuelas, William Yang Wang

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.

📄 PDF Abstract BibTeX arXiv:2407.06426

Code (1)

lukeyoffe/debunc 공식 구현 pytorch

Tasks

Language ModelingLanguage ModellingLarge Language Model

Methods 이 논문이 사용한 방법론

Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…

Attention 설명 없음

Every Response Counts: Quantifying Uncertainty of LLM-based Multi-Agent Systems through Tensor Decomposition

2026-04-09 · Tiejin Chen, Huaiyuan Yao, Jia Chen, Evangelos E. Papalexakis 외 arxiv

While Large Language Model-based Multi-Agent Systems (MAS) consistently outperform single-agent systems on complex tasks, their intricate interactions introduce critical reliability challenges arising from communication …

From Assumptions to Actions: Turning LLM Reasoning into Uncertainty-Aware Planning for Embodied Agents

2026-02-04 · SeungWon Seo, SooBin Lim, SeongRae Noh, Haneul Kim 외 arxiv

Embodied agents operating in multi-agent, partially observable, and decentralized environments must plan and act despite pervasive uncertainty about hidden objects and collaborators' intentions. Recent advances in applyi…

SMAC-Talk: A Natural Language Extension of the StarCraft Multi-Agent Challenge for Large Language Models

2026-06-02 · Joel Sol, Homayoun Najjaran arxiv

As LLMs become more widely deployed, they are increasingly expected to work alongside other AI agents rather than operating in isolation. Effective coordination in these settings requires agents to communicate, share inf…

Decision Making

The Five Ws of Multi-Agent Communication: Who Talks to Whom, When, What, and Why -- A Survey from MARL to Emergent Language and LLMs

2026-02-12 · Jingdi Chen, Hanqing Yang, Zongjun Liu, Carlee Joe-Wong arxiv

Multi-agent sequential decision-making powers many real-world systems, from autonomous vehicles and robotics to collaborative AI assistants. In dynamic, partially observable environments, communication is often what redu…

Multi-agent Reinforcement Learning Autonomous Vehicles

Learning to request guidance in emergent language

2019-11-01 · WS 2019 11 · Benjamin Kolb, Leon Lang, Henning Bartsch, Arwin Gansekoele 외

Previous research into agent communication has shown that a pre-trained guide can speed up the learning process of an imitation learning agent. The guide achieves this by providing the agent with discrete messages in an …

Imitation Learning