Unmediated AI-Assisted Scholarly Citations
Stefan Szeider
TL;DR
The paper tackles the problem of fabrications in AI-assisted citations by decoupling natural-language interaction from data retrieval using the Model Context Protocol (MCP). It presents MCP-DBLP, a backend that provides conversational search and an unmediated BibTeX export, ensuring citations originate directly from authoritative sources. Evaluation across 104 obfuscated citations shows an 82.7% perfect-match rate with zero metadata corruption for unmediated exports, outperforming a Web baseline. The approach enables AI-assisted scholarly writing and autonomous agents while maintaining scholarly integrity and is adaptable to other bibliographic databases.
Abstract
Traditional bibliography databases require users to navigate search forms and manually copy citation data. Language models offer an alternative: a natural-language interface where researchers write text with informal citation fragments, which are automatically resolved to proper references. However, language models are not reliable for scholarly work as they generate fabricated (hallucinated) citations at substantial rates. We present an architectural approach that combines the natural-language interface of LLM chatbots with the accuracy of direct database access, implemented through the Model Context Protocol. Our system enables language models to search bibliographic databases, perform fuzzy matching, and export verified entries, all through conversational interaction. A key architectural principle bypasses the language model during final data export: entries are fetched directly from authoritative sources, with timeout protection, to guarantee accuracy. We demonstrate this approach with MCP-DBLP, a server providing access to the DBLP computer science bibliography. The system transforms form-based bibliographic services into conversational assistants that maintain scholarly integrity. This architecture is adaptable to other bibliographic databases and academic data sources.