Enhancing Multi-Agent Consensus through Third-Party LLM Integration: Analyzing Uncertainty and Mitigating Hallucinations in Large Language Models

Zhihua Duan,Jialin Wang

2024-11-25

Abstract:Large Language Models (LLMs) still face challenges when dealing with complex reasoning tasks, often resulting in hallucinations, which limit the practical application of LLMs. To alleviate this issue, this paper proposes a new method that integrates different LLMs to expand the knowledge boundary, reduce dependence on a single model, and promote in-depth debate among agents. The main contributions include: 1) Introducing third-party LLMs to adjust the attention weights of agents through uncertainty estimation and confidence analysis, optimizing consensus formation in multi-agent systems; 2) Experiments on arithmetic datasets have validated the effectiveness of the method, surpassing traditional multi-agent baselines. This research provides a new perspective for large models to alleviate hallucination phenomena when dealing with complex tasks.

Artificial Intelligence,Computation and Language,Multiagent Systems

What problem does this paper attempt to address?

The problem that this paper attempts to solve is the hallucination phenomenon in large - language models (LLMs) when handling complex reasoning tasks, that is, the content generated by the model does not conform to the facts, which limits the practical applications of LLMs. To alleviate this problem, the paper proposes a new method to expand the knowledge boundary by integrating different LLMs, reduce the dependence on a single model, and promote in - depth debates among agents. Specifically, the main contributions of the paper include:

1. **Introducing third - party LLMs**: Adjust the attention weights of agents through uncertainty estimation and confidence analysis to optimize the consensus - forming process in multi - agent systems.
2. **Experimental verification**: Experiments on arithmetic datasets have verified the effectiveness of this method, surpassing the traditional multi - agent baseline methods.

By introducing third - party LLMs and combining the optimization of uncertainty estimation and attention mechanisms, the paper aims to improve the decision - making ability and problem - solving depth of multi - agent systems when handling complex tasks, thereby reducing the hallucination phenomenon and enhancing the reliability and practicality of the model.