用小波变换分解文本，提升长文档摘要的准确性和真实性。
DWTSumm: Discrete Wavelet Transform for Document Summarization
- 将文本视为语义信号，通过小波变换分离全局结构与局部细节。
- 在临床和法律数据集上，语义相似度提升超2%，事实一致性提高4%以上。
- 适合需要高准确率的医疗、法律等专业领域摘要任务。
使用大语言模型（LLMs）总结长篇领域特定文档仍具挑战性，受限于上下文长度、信息丢失和幻觉问题，尤其在临床和法律场景中。我们提出一种基于离散小波变换（DWT）的多分辨率框架，将文本视为语义信号，分解为全局（近似）和局部（细节）成分。该方法应用于句级或词级嵌入，生成紧凑表示，保留整体结构与关键领域细节，可直接作为摘要或引导LLM生成。在临床和法律基准测试中，性能达到与基准相当的ROUGE-L分数。相较于GPT-4o基线，基于DWT的摘要在语义相似度和事实对齐性方面持续提升：BERTScore提升超2%，语义保真度提升超4%，法律任务中的事实一致性显著增强，且METEOR得分大幅改善，表明领域特有语义得以更好保留。在多种嵌入模型下，保真度最高达97%，说明DWT具有语义去噪作用，减少幻觉并增强事实依据。总体而言，DWT提供了一种轻量、通用的可靠长文档与领域特定摘要方法。
原文摘要 · Abstract (English)
Summarizing long, domain-specific documents with large language models (LLMs) remains challenging due to context limitations, information loss, and hallucinations, particularly in clinical and legal settings. We propose a Discrete Wavelet Transform (DWT)-based multi-resolution framework that treats text as a semantic signal and decomposes it into global (approximation) and local (detail) components. Applied to sentence- or word-level embeddings, DWT yields compact representations that preserve overall structure and critical domain-specific details, which are used directly as summaries or to guide LLM generation. Experiments on clinical and legal benchmarks demonstrate comparable ROUGE-L scores. Compared to a GPT-4o baseline, the DWT based summarization consistently improve semantic similarity and grounding, achieving gains of over 2% in BERTScore, more than 4\% in Semantic Fidelity, factual consistency in legal tasks, and large METEOR improvements indicative of preserved domain-specific semantics. Across multiple embedding models, Fidelity reaches up to 97%, suggesting that DWT acts as a semantic denoising mechanism that reduces hallucinations and strengthens factual grounding. Overall, DWT provides a lightweight, generalizable method for reliable long-document and domain-specific summarization with LLMs.
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。