Dataset Description

This is a VQA dataset based on figures extracted from arXiv publications taken from ArXiVQA dataset from Multimodal ArXiV.

Load the dataset

from datasets import load_dataset
import csv

def load_beir_qrels(qrels_file):
    qrels = {}
    with open(qrels_file) as f:
        tsvreader = csv.DictReader(f, delimiter="\t")
        for row in tsvreader:
            qid = row["query-id"]
            pid = row["corpus-id"]
            rel = int(row["score"])
            if qid in qrels:
                qrels[qid][pid] = rel
            else:
                qrels[qid] = {pid: rel}
    return qrels

corpus_ds = load_dataset("openbmb/VisRAG-Ret-Test-ArxivQA", name="corpus", split="train")
queries_ds = load_dataset("openbmb/VisRAG-Ret-Test-ArxivQA", name="queries", split="train")

qrels_path = "xxxx" # path to qrels file which can be found under qrels folder in the repo.
qrels = load_beir_qrels(qrels_path)