arXiv cs.CL· Ahmet Yavuz Uluslu, Mark Gales, Kate Knill, Gerold Schneider·· 12 小时前AI 评分22
利用 LLM Token 概率从 L2 英语推断母语背景
Inferring L1 Backgrounds from L2 English with LLM Token Probabilities
AI 导读
研究提出一种 Token 概率方法，将文本按指定候选母语的提示词前缀计算续写得分，取最高分作为预测结果，用于从 L2 英语推断说话者的母语背景。
正文 · 原文
Abstract:Native language identification (NLI) infers a non-native speaker's first language (L1) from their second language (L2) production. We investigate a token probability approach that scores each text sample as a continuation of a prompt prefix specifying a candidate L1, then predicts the candidate with the highest score. We evaluate our approach using Llama 3.1 8B and 70B on three datasets: TOEFL9, Write and Improve 2024, and BULATS-L1. We compare its performance with previous prompting-based classification approaches. Token-probability scoring substantially improves on direct classification, particularly with the 8B model, while the 70B model achieves state-of-the-art performance across all datasets. Our results indicate that pretrained LLMs' token probabilities contain information useful for distinguishing L1 backgrounds from L2 English production.
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.23309 [cs.CL] |
| (or arXiv:2610.23309v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.23309 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Ahmet Yavuz Uluslu [view email]
[v1]
Fri, 16 Oct 2026 11:12:42 UTC (300 KB)
来源：arXiv cs.CL · arxiv.org