@inproceedings{feng-etal-2025-cypherbench,
    title = "{C}ypher{B}ench: Towards Precise Retrieval over Full-scale Modern Knowledge Graphs in the {LLM} Era",
    author = "Feng, Yanlin  and
      Papicchio, Simone  and
      Rahman, Sajjadur",
    editor = "Che, Wanxiang  and
      Nabende, Joyce  and
      Shutova, Ekaterina  and
      Pilehvar, Mohammad Taher",
    booktitle = "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2025",
    address = "Vienna, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingestion-acl-25/2025.acl-long.438/",
    pages = "8934--8958",
    ISBN = "979-8-89176-251-0",
    abstract = "Retrieval from graph data is crucial for augmenting large language models (LLM) with both open-domain knowledge and private enterprise data, and it is also a key component in the recent GraphRAG system (CITATION). Despite decades of research on knowledge graphs and knowledge base question answering, leading LLM frameworks (Langchain and LlamaIndex) have only minimal support for retrieval from modern encyclopedic knowledge graphs like Wikidata. In this paper, we analyze the root cause and suggest that modern RDF knowledge graphs (Wikidata, Freebase) are less efficient for LLMs due to overly large schemas that far exceed the typical LLM context window, use of resource identifiers, overlapping and ambiguous relation types and lack of normalization. As a solution, we propose \textit{property graph views} on top of the underlying RDF graph that can be efficiently queried by LLMs using \textit{Cypher}. We instantiated this idea on Wikidata and introduced CypherBench, the first benchmark with 11 large-scale, multi-domain property graphs with 7.8 million entities and over 10,000 questions. To achieve this, we tackled several key challenges, including developing an RDF-to-property graph conversion engine, creating a systematic pipeline for text-to-Cypher task generation, and designing new evaluation metrics."
}