arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2410.04068cs.CLcs.AI
收藏

ECon:论证据冲突的检测与消解

ECon: On the Detection and Resolution of Evidence Conflicts

The Hong Kong University of Science and Technology(香港科技大学);Westlake University(西湖大学);Shanghai Jiaotong University(上海交通大学);Amazon AWS AI(亚马逊AWS人工智能)

Cheng Jiayang, Chunkit Chan, Qianqian Zhuang, Lin Qiu, Tianhang Zhang, Tengxiao Liu, Yangqiu Song, Yue Zhang, Pengfei Liu, Zheng Zhang

AI总结:

本研究针对大语言模型兴起带来的证据间冲突问题,提出ECon方法生成多样化经验证的证据冲突以模拟现实场景,评估了NLI、FC模型及LLM的冲突检测效果,分析了LLM的冲突消解行为并得出多项关键发现。

摘要:

大语言模型(LLM)的兴起显著影响了决策系统中的信息质量,使得AI生成内容大行其道,也给错误信息检测与冲突信息(即“证据间冲突”)管理带来了挑战。本研究提出了一种生成多样化、经验证的证据冲突的方法,以模拟现实世界中的错误信息场景。我们针对这些冲突评估了包括自然语言推理(NLI)模型、事实一致性(FC)模型和LLM在内的冲突检测方法(研究问题RQ1),并分析了LLM的冲突消解行为(研究问题RQ2)。我们的核心发现包括:(1)NLI和LLM模型在检测答案冲突时表现出高精度,但较弱的模型存在召回率低的问题;(2)FC模型难以处理词汇相似的答案冲突,而NLI和LLM模型在这方面表现更好;(3)GPT-4等更强的模型表现稳健,尤其在处理细微冲突时。在冲突消解方面,LLM通常会在没有正当理由的情况下偏向某一条冲突证据,且若其存在先验信念则会依赖内部知识。

英文摘要:

The rise of large language models (LLMs) has significantly influenced the quality of information in decision-making systems, leading to the prevalence of AI-generated content and challenges in detecting misinformation and managing conflicting information, or "inter-evidence conflicts." This study introduces a method for generating diverse, validated evidence conflicts to simulate real-world misinformation scenarios. We evaluate conflict detection methods, including Natural Language Inference (NLI) models, factual consistency (FC) models, and LLMs, on these conflicts (RQ1) and analyze LLMs' conflict resolution behaviors (RQ2). Our key findings include: (1) NLI and LLM models exhibit high precision in detecting answer conflicts, though weaker models suffer from low recall; (2) FC models struggle with lexically similar answer conflicts, while NLI and LLM models handle these better; and (3) stronger models like GPT-4 show robust performance, especially with nuanced conflicts. For conflict resolution, LLMs often favor one piece of conflicting evidence without justification and rely on internal knowledge if they have prior beliefs.

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。

↑