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

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

ScholarCopilot: Training Large Language Models for Academic Writing with Accurate Citations: 43 upvotes on Hugging Face Daily Papers, #5 of 22 papers on 2025-04-03. Day-by-day upvote history.

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.

Paper page on Hugging Face · arXiv

Data: hysts-bot-data/daily-papers-stats and the Daily Papers API. Open data: tardellirs/paper-pulse-data. Sister project: Model Pulse, the download history of every model on the Hub.