arXiv cs.CL· Bastian Bunzeck, Francesca Padovani·· 15 小时前AI 评分22
研究：合成儿童-成人对话在发育上不具合理性
Synthetic Child-Adult Interactions Are Not Developmentally Plausible
AI 导读
一项 arXiv 研究对比合成 TinyDialogues 数据集与自然对话后发现，合成数据在词汇层面与自然数据相当，却严重低估了口语交互碎片化、杂乱的特征。研究提出用真实对话数据做指令微调，初步实验显示小模型经适度参数高效微调后可接近自然数据的统计倾向。
正文 · 原文
Abstract:Synthetic data has become a common resource for training language models, and this trend has spilled over into research on developmentally plausible language modeling. However, it remains unclear whether LM-generated child-adult interactions actually resemble natural ones. We compare the synthetic TinyDialogues dataset to natural dialogue across a broad range of linguistic measures. While synthetic data matches natural data on the lexical level, it drastically underestimates the messy, fragmented nature of spoken interaction. To mitigate this, we propose instruction-tuning with data created from real dialogues, and show in a pilot study that a small model approximates the statistical tendencies of natural data after moderate parameter-efficient fine-tuning.
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.35029 [cs.CL] |
| (or arXiv:2610.35029v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.35029 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Francesca Padovani [view email]
[v1]
Mon, 26 Oct 2026 08:40:54 UTC (9,965 KB)
来源：arXiv cs.CL · arxiv.org