用思维链减少大模型新闻时间线总结中的幻觉
NTS-CoT: Mitigating Hallucinations in LLM-based News Timeline Summarization with Chain-of-Thought Reasoning
- 通过思维链捕捉新闻关键要素，保证摘要忠实度
- 结合时间重要性和事件显著性选时间戳，避免遗漏
- 推理因果关系，解决事件时间线中的信息缺失问题
在线新闻更新迅速，事件追踪困难，亟需时间线摘要（TLS）。现有基于大模型的TLS仍存在幻觉问题，即生成内容偏离原始新闻，且未被充分研究。本文识别出两类主要幻觉：摘要内容失真与时间-事件关联中的信息遗漏。为此提出NTS-CoT框架，利用思维链（CoT）推理缓解幻觉。该框架包含三个模块：1）Element-CoT，提取关键新闻要素以保障摘要忠实性；2）Date Selection，融合时间显著性与事件突出性进行时间戳选择；3）Causal-CoT，推理事件因果关系，减少时间线中信息遗漏。在三个TLS基准上的定量分析与人工评估均表明，NTS-CoT优于现有先进方法，有效降低幻觉并提升性能。代码已公开。
原文摘要 · Abstract (English)
The rapid updates of online news make tracking event developments challenging, highlighting the need for timeline summarization (TLS). Hallucinations, where LLM-generated content deviates from source news, still remain a critical issue in LLM-based TLS and are not well studied in existing works. To bridge this gap, we identify two primary types of hallucinations: unfaithful content during news summarization and information omission in date-event summarization. Then, we propose NTS-CoT, a novel framework that leverages Chain-of-Thought (CoT) reasoning to mitigate hallucinations in TLS. The framework consists of three key modules: i) Element-CoT to capture essential news elements for faithful summarization, ii) Date Selection to combine temporal saliency and event prominence for timestamp selection, and iii) Causal-CoT to infer causal relationships and reduce omissions in date-event summarization. Extensive experiments, including quantitative analysis on three TLS benchmarks and human evaluation, demonstrate that NTS-CoT outperforms state-of-the-art baselines, effectively mitigating hallucinations and improving LLM-based TLS performance. Our source code is available at https://anonymous.4open.science/r/NTS-CoT .
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。