用原文上下文增强长文档摘要，减少大模型幻觉。
Context-Aware Hierarchical Merging for Long Document Summarization
- 用原文内容替换或修正中间摘要，提升信息准确性
- 在法律和叙事数据集上，优于零样本与基础合并方法
- 结合抽取式摘要效果更佳，适合长文本生成任务
层次化合并常用于处理超长文本（>100K tokens），通过分段摘要再递归合并生成最终摘要。尽管缓解了大语言模型（LLMs）输入长度限制，但递归合并易放大幻觉，导致事实错误。本文提出通过引入源文档上下文来减轻幻觉，包括：用相关原文替换中间摘要、以上下文为依据修正摘要、以及通过引用隐式对齐摘要与原文。在法律和叙事领域数据集上的实验表明，该方法在Llama 3.1模型上持续优于零样本及基础层次合并基线。分析显示，结合抽取式摘要识别相关文本时，修正类方法表现最佳。
原文摘要 · Abstract (English)
Hierarchical Merging is a technique commonly used to summarize very long texts ($>$100K tokens) by breaking down the input into smaller sections, summarizing those sections individually, and then merging or combining those summaries into a final coherent summary. Although it helps address the limitations of large language models (LLMs) with fixed input length constraints, the recursive merging process can amplify LLM hallucinations, increasing the risk of factual inaccuracies. In this paper, we seek to mitigate hallucinations by enriching hierarchical merging with context from the source document. Specifically, we propose different approaches to contextual augmentation ranging from \emph{replacing} intermediate summaries with relevant input context, to \emph{refining} them while using the context as supporting evidence, and \emph{aligning} them implicitly (via citations) to the input. Experimental results on datasets representing legal and narrative domains show that contextual augmentation consistently outperforms zero-shot and hierarchical merging baselines for the Llama 3.1 model family. Our analysis further reveals that refinement methods tend to perform best when paired with extractive summarization for identifying relevant input.
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。