OUNLP在TSAR 2025共享任务中的表现：通过代码生成的多轮文本简化
OUNLP at TSAR 2025 Shared Task: Multi-Round Text Simplifier via Code Generation
AI总结：
OUNLP通过代码生成在TSAR 2025共享任务中提出多轮文本简化方法，利用基于规则和LLM的联合简化技术提升可读性控制的文本简化性能。
摘要：
本文描述了提交至TSAR-2025共享任务（Alva-Manchego等人，2025）的OUNLP系统，该系统旨在使用基于大语言模型提示的生成进行可控可读性文本简化。基于对基于提示的文本简化方法的分析，我们发现文本简化性能高度依赖于源CEFR（Arase等人，2022）级别与目标CEFR级别之间的差距。受这一发现的启发，我们提出了两种多轮简化方法，并通过GPT-4o生成了它们：基于规则的简化（MRS-Rule）和联合基于规则的LLM简化（MRS-Joint）。我们提交的系统在20个团队中排名第7。后来通过MRS-Joint的改进表明，以LLM简化候选文本为起点可以进一步提升多轮简化性能。
英文摘要：
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.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
