使用大语言模型生成不同可读性水平的教育材料
Generating Educational Materials with Different Levels of Readability using LLMs
AI总结:
本研究提出分级文本生成任务,评估三款大模型通过零样本和少样本提示生成不同可读性教育材料的能力,发现少样本提示可提升性能,同时指出存在错误信息等问题,需进一步研究保障内容质量。
摘要:
本研究提出分级文本生成任务,目标是在保留原意的前提下将教育材料改写为特定可读性水平。我们评估了GPT-3.5、LLaMA-2 70B和Mixtral 8x7B通过零样本和少样本提示生成不同可读性水平内容的能力。对100份处理后的教育材料的评估显示,少样本提示显著提升了可读性调控和信息保留的性能。LLaMA-2 70B在达到目标难度范围方面表现更优,GPT-3.5则更能保留原意。不过人工检查发现了引入错误信息、编辑分布不一致等问题。这些发现表明需进一步研究以保障生成教育内容的质量。
英文摘要:
This study introduces the leveled-text generation task, aiming to rewrite educational materials to specific readability levels while preserving meaning. We assess the capability of GPT-3.5, LLaMA-2 70B, and Mixtral 8x7B, to generate content at various readability levels through zero-shot and few-shot prompting. Evaluating 100 processed educational materials reveals that few-shot prompting significantly improves performance in readability manipulation and information preservation. LLaMA-2 70B performs better in achieving the desired difficulty range, while GPT-3.5 maintains original meaning. However, manual inspection highlights concerns such as misinformation introduction and inconsistent edit distribution. These findings emphasize the need for further research to ensure the quality of generated educational content.
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。