ScholarCopilot：訓練大型語言模型以實現精確引用的學術寫作

ScholarCopilot: Training Large Language Models for Academic Writing with Accurate Citations

April 1, 2025

作者: Yubo Wang, Xueguang Ma, Ping Nie, Huaye Zeng, Zhiheng Lyu, Yuxuan Zhang, Benjamin Schneider, Yi Lu, Xiang Yue, Wenhu Chen

cs.AI

摘要

學術寫作既需要連貫的文本生成，又要求精確引用相關文獻。儘管近期的檢索增強生成（RAG）系統在通用文本生成的事實準確性上取得了顯著提升，但其在支持專業學術寫作方面的能力仍顯不足。本研究提出了ScholarCopilot，這是一個統一框架，旨在增強現有大型語言模型，以生成具有準確且上下文相關引用的專業學術文章。ScholarCopilot通過生成檢索標記[RET]來動態決定何時檢索學術參考文獻，並利用其表徵從數據庫中查找相關引用。檢索到的參考文獻被輸入模型以增強生成過程。我們在單一框架內聯合優化生成與引用任務，以提高效率。在arXiv的50萬篇論文上訓練後，我們的模型在評估數據集上達到了40.1%的Top-1檢索準確率，超越了如E5-Mistral-7B-Instruct（15.0%）和BM25（9.8%）等基線模型。在1000份學術寫作樣本的數據集上，ScholarCopilot在生成質量（涵蓋相關性、連貫性、學術嚴謹性、完整性和創新性）上獲得16.2/25分，超越了參數量多出10倍的模型如Qwen-2.5-72B-Instruct（15.8/25）。人體研究也證實了ScholarCopilot在引用召回率、寫作效率和整體用戶體驗上的卓越表現，驗證了我們方法的有效性。

English

Academic writing requires both coherent text generation and precise citation of relevant literature. Although recent Retrieval-Augmented Generation (RAG) systems have significantly improved factual accuracy in general-purpose text generation, their capacity to adequately support professional academic writing remains limited. In this work, we introduce ScholarCopilot, a unified framework designed to enhance existing large language models for generating professional academic articles with accurate and contextually relevant citations. ScholarCopilot dynamically determines when to retrieve scholarly references by generating a retrieval token [RET], and then utilizes its representation to look up relevant citations from a database. The retrieved references are fed into the model to augment the generation process. We jointly optimize both the generation and citation tasks within a single framework to increase efficiency. Trained on 500K papers from arXiv, our model achieves a top-1 retrieval accuracy of 40.1% on our evaluation dataset, outperforming baselines such as E5-Mistral-7B-Instruct (15.0%) and BM25 (9.8%). On a dataset of 1,000 academic writing samples, ScholarCopilot scores 16.2/25 in generation quality (measured across relevance, coherence, academic rigor, completeness, and innovation), surpassing models with 10x more parameters such as Qwen-2.5-72B-Instruct (15.8/25). Human studies also confirm ScholarCopilot's superior performance in citation recall, writing efficiency, and overall user experience, confirming the effectiveness of our approach.