README.md

metadata

license: mit
task_categories:
  - table-question-answering
language:
  - en
tags:
  - rag
  - tables
  - multi-table
  - question-answering
  - fact-checking
  - llm
configs:
  - config_name: table
    data_files: tabfact_table.jsonl
  - config_name: test_query
    data_files: tabfact_query.jsonl

📄 RAG over Tables: Hierarchical Memory Index, Multi-Stage Retrieval, and Benchmarking | 👨🏻‍💻 Code | 🤗 MultiTableQA Hub Collection

This repository contains MultiTableQA-TabFact, one of the five datasets released as part of the comprehensive MultiTableQA benchmark. MultiTableQA-TabFact focuses on table fact-checking tasks. The overall MultiTableQA benchmark extends the traditional single-table QA setting into a multi-table retrieval and question answering benchmark, enabling more realistic and challenging evaluations.

Other datasets in the MultiTableQA benchmark include:

Dataset	Link
MultiTableQA-TATQA	🤗 dataset link
MultiTableQA-TabFact	🤗 dataset link
MultiTableQA-SQA	🤗 dataset link
MultiTableQA-WTQ	🤗 dataset link
MultiTableQA-HybridQA	🤗 dataset link

---

Sample Usage

To download and preprocess the MultiTableQA benchmark, navigate to the table2graph directory in the code repository and run the prepare_data.sh script:

cd table2graph
bash scripts/prepare_data.sh

This script will automatically fetch the source tables, apply decomposition (row/column splitting), and generate the benchmark splits.

---

Citation

If you find our work useful, please cite:

@misc{zou2025rag,
      title={RAG over Tables: Hierarchical Memory Index, Multi-Stage Retrieval, and Benchmarking}, 
      author={Jiaru Zou and Dongqi Fu and Sirui Chen and Xinrui He and Zihao Li and Yada Zhu and Jiawei Han and Jingrui He},
      year={2025},
      eprint={2504.01346},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2504.01346}, 
}