This dataset is part of a Table + Text retrieval benchmark. Includes queries and relevance judgments across test split(s), with corpus in 1 format(s): corpus.

Changelog

[2026-05-10] Updated queries.jsonl with corrected annotations. Details: Fixed [incorrect query annotations and symbolic errors]. Version: Full diff available at ea30775

	Configs

Config Description Split(s)

default Relevance judgments (qrels): qid, did, score test… See the full description on the dataset page: https://huggingface.co/datasets/ibm-research/AITQARetrieval.