A Co-design Framework of Event-Centric Knowledge Graphs and Agentic Retrieval-Augmented Generation for Multi-Hop Question Answering over Engineering Heritage Archives
Abstract
This repository contains the evaluation assets for the paper "A Co-design Framework of Event-Centric Knowledge Graphs and Agentic Retrieval-Augmented Generation for Multi-Hop Question Answering over Engineering Heritage Archives" (2026). The assets cover two QA benchmarks: BridgeQA — a self-constructed multi-hop QA dataset of 1,088 question-answer pairs derived from 36 archival documents on Chinese bridge culture and heritage. MultiHop-RAG — a public multi-hop QA benchmark of 2,556 questions (Tang & Yang, 2024). For each benchmark, the repository provides: (a) the raw prediction outputs of every model evaluated in the paper, (b) a self-contained evaluation pipeline, (c) pre-computed evaluation results that can be inspected directly without running any code, and (d) the parameter sensitivity data for the disambiguation component of EvAgentRAG. All assets are released under CC BY 4.0 to enable independent reproduction and verification.
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.