cs.CL提交于 2026-10-08已译

使用大语言模型实现爱沙尼亚语文档级文本简化

Document-Level Text Simplification in Estonian Using Large Language Models

Meeri-Ly Muru · Eduard Barbu

中文摘要

文档级文本简化涉及超越句子内部编辑的转换,涵盖篇章连贯性(coherence)、回指消解(anaphora resolution)与跨段落一致性(cross-paragraph consistency)。尽管针对高资源语言的句子级简化已取得进展,但在形态丰富且低资源的语言(如爱沙尼亚语)中,文档级简化仍鲜有探索。本研究对五个当前最优的多语言大语言模型(LLM)在爱沙尼亚语文档级简化任务上的表现进行全面评估,考察了三种提示策略:单遍生成(single-pass generation)、基于管道的模块化智能体(pipeline-based modular agents)以及加入指南增强的管道。评估框架整合了衡量可读性、语义保持与篇章连贯性的自动指标,并辅以结构化的人工标注协议。研究表明,Gemini-2.0 与 LLaMA-3.3 生成的输出接近母语水平且语义保持良好,而其他模型则存在显著的语法与语义缺陷。本工作的贡献包括新颖的文档级连贯性指标、基于证据的提示策略,以及面向可复现性的公开资源。

关键要点

01问题:形态丰富且低资源的爱沙尼亚语在文档级文本简化方向上几乎未被研究,缺乏系统评估
02方法:对五个多语言 LLM 评估三种提示策略(单遍生成、模块化管道、指南增强管道),结合自动指标与人工标注
03结果:Gemini-2.0 与 LLaMA-3.3 在流利度与语义保持上接近母语水平,其余模型存在语法与语义缺陷
04贡献:提出新的文档级连贯性指标、循证的提示策略,以及可复现的公开资源

解读

尚无解读。

原始英文摘要

arXiv:2610.10378v1 Announce Type: new Abstract: Document-level text simplification involves transformations that go beyond sentence-internal edits, addressing discourse coherence, anaphora resolution, and cross-paragraph consistency. Despite advances in sentence-level simplification for high-resource languages, document-level simplification in morphologically rich, low-resource languages such as Estonian remains largely unexplored. This study presents a comprehensive evaluation of five state-of-the-art multilingual large language models (LLMs) for document-level simplification in Estonian. Three prompting strategies are examined: single-pass generation, pipeline-based modular agents, and guideline-augmented pipelines. The evaluation framework integrates automatic metrics assessing readability, semantic preservation, and discourse coherence, alongside a structured manual annotation protocol. The findings indicate that Gemini-2.0 and LLaMA-3.3 produce outputs with near-native fluency and strong meaning preservation, whereas other models display notable grammatical and semantic limitations. This work contributes novel document-level coherence metrics, evidence-based prompting strategies, and publicly available resources for reproducibility.