MATEval: A Multi-Agent Discussion Framework for Advancing Open-Ended Text Evaluation
Yu Li, Shenyu Zhang, Rui Wu, Xiutian Huang, Yongrui Chen, Wenhao Xu, Guilin Qi, Dehai Min
TL;DR
Open-ended text evaluation with LLMs suffers from instability and misalignment with human judgments. MATEval introduces a Multi-Agent Text Evaluation framework where GPT-4-based Evaluator, Feedback, and Summarizer agents engage in self-reflection and Chain-of-Thought guided discussions, with a feedback loop and two report formats. The approach yields higher correlation with human judgments than existing methods across English and Chinese storytelling datasets, including an Alipay industrial case, and ablation confirms the value of feedback, explanations, and multi-agent collaboration. The work enables more reliable error localization, faster model iteration, and practical deployment in industry, with future directions toward domain-specific domain-tuned agents and cross-domain collaboration.
Abstract
Recent advancements in generative Large Language Models(LLMs) have been remarkable, however, the quality of the text generated by these models often reveals persistent issues. Evaluating the quality of text generated by these models, especially in open-ended text, has consistently presented a significant challenge. Addressing this, recent work has explored the possibility of using LLMs as evaluators. While using a single LLM as an evaluation agent shows potential, it is filled with significant uncertainty and instability. To address these issues, we propose the MATEval: A "Multi-Agent Text Evaluation framework" where all agents are played by LLMs like GPT-4. The MATEval framework emulates human collaborative discussion methods, integrating multiple agents' interactions to evaluate open-ended text. Our framework incorporates self-reflection and Chain-of-Thought (CoT) strategies, along with feedback mechanisms, enhancing the depth and breadth of the evaluation process and guiding discussions towards consensus, while the framework generates comprehensive evaluation reports, including error localization, error types and scoring. Experimental results show that our framework outperforms existing open-ended text evaluation methods and achieves the highest correlation with human evaluation, which confirms the effectiveness and advancement of our framework in addressing the uncertainties and instabilities in evaluating LLMs-generated text. Furthermore, our framework significantly improves the efficiency of text evaluation and model iteration in industrial scenarios.