通过多轮代码生成提升文本可读性，基于CEFR等级差优化简化效果
OUNLP at TSAR 2025 Shared Task: Multi-Round Text Simplifier via Code Generation
- 利用源文与目标文的CEFR等级差设计多轮简化策略
- 基于GPT-4o生成规则与联合简化方案，提升可读性控制精度
- 适合需要精细可读性调节的自然语言生成任务
本文介绍OU NLP团队在TSAR-2025共享任务中的系统，采用基于大模型提示的生成方法实现可控可读性文本简化。通过对提示式文本简化方法的分析，发现简化效果与源文本和目标文本之间的CEFR等级差距密切相关。受此启发，我们提出两种多轮简化方法：基于规则的简化（MRS-Rule）和规则与大模型联合简化（MRS-Joint），并通过GPT-4o进行生成。所提交系统在20支队伍中排名第七。后续改进显示，以大模型生成的简化候选为起点，可进一步提升多轮简化性能。
原文摘要 · Abstract (English)
This paper describes the OUNLP system submitted to the TSAR-2025 Shared Task (Alva-Manchego et al., 2025), designed for readability-controlled text simplification using LLM-prompting-based generation. Based on the analysis of prompt-based text simplification methods, we discovered an interesting finding that text simplification performance is highly related to the gap between the source CEFR (Arase et al., 2022) level and the target CEFR level. Inspired by this finding, we propose two multi-round simplification methods and generate them via GPT-4o: rule-based simplification (MRS-Rule) and jointly rule-based LLM simplification (MRS-Joint). Our submitted systems ranked 7 out of 20 teams. Later improvements with MRS-Joint show that taking the LLM simplified candidates as the starting point could further boost the multi-round simplification performance.
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。