arXiv cs.CL· Wenshi Gu·· 13 小时前AI 评分8
基于 RAG+Prompt 的日汉翻译系统：面向名词修饰从句的循证翻译研究（已撤稿）
From prompting to evidence-based translation: A RAG+prompt system for Japanese-Chinese translation and its pedagogical potential
AI 导读
一项已被作者撤稿的研究提出 RAG+Prompt 日汉翻译系统，在不修改基座模型的前提下结合语言学分析与嵌入检索，用 GPT-4o 在 66 句测试集上评测。知识库从 0 增至 2,000 条时，句级宏平均 BLEU 由 24.28 升至 29.96（+5.68，+23.4%）。作者因发现实验数据存在错误而撤回该版本。
正文 · 原文
This paper has been withdrawn by Wenshi Gu
No PDF available, click to view other formats
Abstract:Large language models perform well on high-resource pairs but are less reliable for Japanese-Chinese sentences containing noun-modifying clause constructions (NMCCs). This study evaluates a retrieval-augmented generation RAG+Prompt translation system that integrates linguistic analysis, embedding-based retrieval, prompt construction, and LLM generation without modifying the base model. The analysis module outputs A1 (inner vs. outer NMCC) and A2 (risk predictions: lexical choice/NMCC handling/word order/style/register); top-k = 5 similar Ja-Zh examples (L2 distance) and A1/A2 are inserted into an enhanced prompt. Using GPT-4o and a 66-sentence test set, we compare six knowledge-base sizes (0/100/200/500/1,000/2,000). Macro-averaged sentence-level BLEU (1-4-gram with brevity penalty; cased; Chinese at the character level) is the sole metric. Mean BLEU increases from 24.28 at 0 (RAG disabled) to 29.96 at 2,000 (+5.68; +23.4%). The upward trend holds across sizes, with larger knowledge bases yielding higher scores. We conclude that the RAG+Prompt translation system improves Ja-Zh translation of sentences containing NMCCs in an interpretable and auditable manner. Limitations include one base model, one metric, and reliance on published texts and commercial APIs; future work will broaden genres, language pairs, and evaluation metrics.
| Comments: | The authors have identified errors in the experimental data and have decided to withdraw this version of the manuscript |
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2605.03387 [cs.CL] |
| (or arXiv:2605.03387v2 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2605.03387 arXiv-issued DOI via DataCite |
Submission history
From: Wenshi Gu [view email]
[v1]
Tue, 5 May 2026 05:48:08 UTC (413 KB)
[v2]
Wed, 21 Oct 2026 06:28:10 UTC (1 KB) (withdrawn)
来源：arXiv cs.CL · arxiv.org