MMD-Flagger:利用最大均值差异检测幻觉
MMD-Flagger: Leveraging Maximum Mean Discrepancy to Detect Hallucinations
AI总结:
针对智能体AI系统中大型语言模型的幻觉问题,提出基于最大均值差异(MMD)的MMD-Flagger方法,通过监控不同解码温度下的输出轨迹检测幻觉,在MUCH基准上用Llama-3、Gemma-3模型完成评估。
摘要:
大型语言模型(LLMs)正越来越多地被集成到智能体AI系统中,但其产生幻觉的倾向仍是一个关键的安全问题。在测试时检测这些事实错误,尤其是在没有真实标签的情况下,对于构建可信的自主智能体至关重要。我们提出MMD-Flagger,一种利用最大均值差异(MMD)并监控LLM在不同解码温度下输出稳定性的幻觉检测方法。我们的方法跟踪LLM在特定解码配置下的响应与一组随机样本之间的MMD轨迹,基于轨迹的特征形状识别幻觉。我们在多语言声明验证基准(MUCH)上使用Llama-3系列和Gemma-3等现代LLM评估了MMD-Flagger。
英文摘要:
Large Language Models (LLMs) are increasingly integrated into agentic AI systems, yet their propensity to generate hallucinations remains a critical safety concern. Detecting these factual errors at test-time, particularly without ground-truth labels, is essential for building trustworthy autonomous agents. We propose MMD-Flagger, an hallucination detection method that utilizes Maximum Mean Discrepancy (MMD) and monitors the stability of LLM outputs across varying decoding temperatures. Our method tracks the MMD trajectory between a LLM's response at a certain decoding configuration and a set of stochastic samples, identifying hallucinations based on the trajectory's characteristic shape. We evaluate MMDFlagger on multi-lingual claim verification benchmarks (MUCH) using modern LLMs like Llama-3 families and Gemma-3.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。