一个SMART助记符听起来像“胶水补药”:将LLM与学生反馈相结合,使助记学习更牢固
A SMART Mnemonic Sounds like "Glue Tonic": Mixing LLMs with Student Feedback to Make Mnemonic Learning Stick
AI总结:
本文提出基于学生表达型与观察型反馈训练和直接偏好优化的SMART助记符生成器,以更低成本达到GPT-4水平并提升学习效果。
摘要:
关键词助记符是将新术语与更简单的关键词联系起来的易记解释。先前的工作为学生生成助记符,但并未使用学生偏好且有助于学习的助记符来训练模型。我们构建了SMART,一个基于真实学生学习新术语时反馈训练的助记符生成器。为了训练SMART,我们首先在一组精心整理的用户编写的助记符上微调LLaMA-2。然后我们使用LLM对齐来增强SMART:我们将SMART生成的助记符部署在一个闪卡应用中,以发现学生偏好的助记符。我们从45名学生中收集了2684条偏好,分为两种类型:表达型(从评分推断)和观察型(从学生学习推断),得出三个关键发现。第一,表达型偏好与观察型偏好不一致;学生认为有帮助的内容并不总能捕捉到真正有帮助的内容。第二,贝叶斯模型可以将来自多种偏好类型的互补数据综合为单一的有效性信号。SMART通过在该信号上进行直接偏好优化来调优,这解决了典型成对比较方法中的平局和缺失标签问题,从而增强了数据以提升LLM输出质量。第三,助记符专家评估SMART在部署成本远低于GPT-4的情况下达到了与GPT-4相当的水平,表明在教育中捕捉多样化学生反馈来对齐LLM的实用性。
英文摘要:
Keyword mnemonics are memorable explanations that link new terms to simpler keywords. Prior work generates mnemonics for students, but they do not train models using mnemonics students prefer and aid learning. We build SMART, a mnemonic generator trained on feedback from real students learning new terms. To train SMART, we first fine-tune LLaMA-2 on a curated set of user-written mnemonics. We then use LLM alignment to enhance SMART: we deploy mnemonics generated by SMART in a flashcard app to find preferences on mnemonics students favor. We gather 2684 preferences from 45 students across two types: expressed (inferred from ratings) and observed (inferred from student learning), yielding three key findings. First, expressed and observed preferences disagree; what students think is helpful does not always capture what is truly helpful. Second, Bayesian models can synthesize complementary data from multiple preference types into a single effectiveness signal. SMART is tuned via Direct Preference Optimization on this signal, which resolves ties and missing labels in the typical method of pairwise comparisons, augmenting data for LLM output quality gains. Third, mnemonic experts assess SMART as matching GPT-4 at much lower deployment costs, showing the utility of capturing diverse student feedback to align LLMs in education.
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。