小模型通过精调提升短文本重写质量，更准确且更少幻觉。
Short-form Text Rewriting with Phi Silica
- 用公开幻灯片数据集+大模型生成标注，构建短文本重写训练集。
- 微调后模型在语义保真度和抗幻觉上显著优于原始小模型。
- 适合需要高精度、低延迟的短文本重写场景，如演示文稿优化。
短文本重写是一种上下文受限、语义密集的改写任务，对模型准确性要求极高。尽管大模型表现良好，小语言模型（SLMs）在此类任务中常因语义失真和幻觉而表现不佳。本文针对小模型Phi Silica，通过数据集构建、提示蒸馏、参数高效微调与评估，系统研究其在短文本重写中的适配。我们从公开幻灯片中提取短篇展示文本，利用GPT-5-chat生成重写监督信号并进行模型评分。实验表明，微调后模型在语义保真度、幻觉减少及用户偏好胜率上均显著提升，优于原版小模型，并接近大模型表现。结果表明，针对性适配可大幅缩小小模型与云端大模型的差距，为高精度重写任务提供实用方案。
原文摘要 · Abstract (English)
Short-form text rewriting is a constrained variant of paraphrasing in which limited context and high semantic density leave little room for variation. While large language models perform well on general paraphrasing, small language models (SLMs) often struggle with semantic fidelity and hallucination robustness in short-form settings. In this work, we present an empirical study of adapting an SLM, Phi Silica, for short-form rewrite through dataset curation, prompt distillation, parameter-efficient fine-tuning, and evaluation. We curate a dataset of short presentation-style text from public slide decks and use GPT-5-chat both to generate rewrite supervision and to conduct LLM-as-a-judge evaluation. Our results show that finetuning improves semantic fidelity, reduces hallucinations, and increases preference win rate against GPT-5-chat rewrites. The findings suggest that targeted adaptation for SLMs can substantially narrow the gap to cloud models and provide practical guidance for adapting SLMs to precision-critical rewrite tasks.
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。