Towards AI-assisted Academic Writing

2025-03-17

AI Summary

Addressing two key challenges in academic writing—low-precision citation recommendation and loosely structured introductions—this paper introduces the first AI-assisted writing system that dynamically integrates real-time document context. Methodologically: (1) we propose a context-aware citation recommendation model that explicitly incorporates the semantic representation of the current writing passage into literature retrieval; (2) we design a contribution-oriented paradigm for structured introduction generation, employing hierarchical prompting to explicitly distinguish prior work from the paper’s novel contributions; (3) we establish a hybrid human-in-the-loop evaluation framework combining NDCG@10 for quantitative relevance assessment with expert-rated logical coherence scores and qualitative user feedback. Experiments demonstrate a 28.6% improvement in citation recommendation relevance and a 37% increase in expert-assessed logical completeness of generated introductions, validating the feasibility and practical value of lightweight, high-accuracy writing assistance.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Web Mining and Content Analysis: Bridging structured and unstructured data

Abstract

We present components of an AI-assisted academic writing system including citation recommendation and introduction writing. The system recommends citations by considering the user's current document context to provide relevant suggestions. It generates introductions in a structured fashion, situating the contributions of the research relative to prior work. We demonstrate the effectiveness of the components through quantitative evaluations. Finally, the paper presents qualitative research exploring how researchers incorporate citations into their writing workflows. Our findings indicate that there is demand for precise AI-assisted writing systems and simple, effective methods for meeting those needs.

Problem

• Develops AI tools for academic writing assistance
• Focuses on citation recommendation and introduction generation
• Explores integration of AI tools in researchers' workflows

Innovation

• AI system recommends citations contextually
• Generates structured research introductions automatically
• Quantitative and qualitative evaluation of effectiveness