While state-of-the-art large language models (LLMs) can excel at adapting text from one style to another, current work does not address the explainability of style transfer models. Recent work has explored generating textual explanations from larger teacher models and distilling them into smaller student models. One challenge with such approach is that LLM outputs may contain errors that require expertise to correct, but gathering and incorporating expert feedback is difficult due to cost and availability. To address this challenge, we propose ICLEF, a novel human-AI collaboration approach to model distillation that incorporates scarce expert human feedback by combining in-context learning and model self-critique. We show that our method leads to generation of high-quality synthetic explainable style transfer datasets for formality (e-GYAFC) and subjective bias (e-WNC). Via automatic and human evaluation, we show that specialized student models fine-tuned on our datasets outperform generalist teacher models on the explainable style transfer task in one-shot settings, and perform competitively compared to few-shot teacher models, highlighting the quality of the data and the role of expert feedback. In an extrinsic task of authorship attribution, we show that explanations generated by smaller models fine-tuned on e-GYAFC are more predictive of authorship than explanations generated by few-shot teacher models.

翻译：尽管最先进的大语言模型（LLM）能够出色地将文本从一种风格适应到另一种风格，但现有研究尚未解决风格迁移模型的可解释性问题。近期研究探索了从较大的教师模型生成文本解释并将其蒸馏到较小的学生模型中的方法。此类方法面临的一个挑战是：LLM输出可能包含需要专业知识才能纠正的错误，但由于成本与可用性限制，收集并整合专家反馈十分困难。为应对这一挑战，我们提出ICLEF——一种新颖的人机协作模型蒸馏方法，该方法通过结合上下文学习与模型自批判机制，将稀缺的人类专家反馈纳入训练过程。我们证明，该方法能够为正式度（e-GYAFC）和主观偏见（e-WNC）生成高质量的可解释风格迁移合成数据集。通过自动评估与人工评估，我们发现在单样本设置下，基于我们数据集微调的专业化学生模型在可解释风格迁移任务上优于通用教师模型，并与少样本教师模型表现相当，这凸显了数据质量与专家反馈的关键作用。在作者归属识别的外部任务中，基于e-GYAFC微调的小模型生成的解释比少样本教师模型生成的解释具有更强的作者身份预测能力。