A Community-driven Hub for High-Quality Text2Cypher Benchmarks and Resources
Abstract
Abstract Property graphs (PGs) are increasingly adopted to represent interconnected data in several application domains. Recent advances in Large Language Models have considerably improved the translation of natural language (NL) questions into executable Cypher queries (usually referred to as the Text2Cypher task). However, the development and evaluation of new Text2Cypher methodologies remain hindered by the limited availability of high-quality benchmark resources and the lack of a common framework for sharing and comparing datasets, graph snapshots, source code, models, and evaluation artifacts. We introduce Text2Cypher-Hub, a community platform designed to facilitate the dissemination, discovery, and reuse of Text2Cypher resources. As the initial contribution to this ecosystem, we release three curated benchmarks---bio2C, twt2C, and fin2C---covering biomedical, social-media, and financial PGs. Together, they comprise 6,500 technically and semantically validated NL--Cypher pairs spanning five complexity categories and capturing realistic domain information needs rather than template-generated query variations. The paper describes the curation workflow adopted to construct the datasets and demonstrates their suitability for supervised fine-tuning and retrieval-augmented generation pipelines. Text2Cypher-Hub complements these benchmarks through a community-driven registry that provides standardized descriptions of Text2Cypher resources while allowing contributors to…
Read the paper
Related studies
- Knowledge Management Perspective of Generative Artificial Intelligence
- Navigating Generative Artificial Intelligence Promises and Perils for Knowledge and Creative Work
- Characterizing the evolving research on enterprise content management
- Large language models and multimodal foundation models for precision oncology
- From Generative AI to Extended Reality: Multidisciplinary Perspectives on the Challenges, Opportunities, and Future of Educational Computing
- Generative Artificial Intelligence: A New Engine for Advancing Environmental Science and Engineering
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.