andreiaalexa/adaptive-rag-scifact-150

SciFact — 150 Queries Sample

A fixed random sample of 150 queries from the SciFact test set, used as the evaluation set for the Adaptive RAG for Token-Efficiency course project.

Dataset Details

Field	Value
Source	BeIR/scifact
Split used	test
Sample size	150 queries
Random seed	0
Domain	Biomedical / Scientific fact-checking

Why this sample?

This 150-query sample covers 50% of the full SciFact test set, offering stronger statistical reliability than the 50-query version while remaining feasible to run locally. Use this for final headline results; use the 50-query version for rapid iteration during development.

The seed is fixed at 0 — every team member gets the same 150 queries.

How to load

from datasets import load_dataset

ds = load_dataset("andreiaalexa/adaptive-rag-scifact-150")
queries = ds["queries"]

print(queries[0])
# {'_id': '...', 'text': '...'}

How it was created

import random
from datasets import load_dataset, DatasetDict

queries_ds = load_dataset("BeIR/scifact", "queries", split="queries")

random.seed(0)
indices = random.sample(range(len(queries_ds)), 150)
sampled = queries_ds.select(indices)

DatasetDict({"queries": sampled}).push_to_hub("andreiaalexa/adaptive-rag-scifact-150")

Project context

This dataset is part of the Adaptive RAG for Token-Efficiency project, developed for the Language Technology / Information Retrieval course. The project investigates whether adaptively selecting the number of retrieved documents per query can reduce token usage without hurting answer quality.

• Full project repository: (link to GitHub repo)
• Smaller sample (50 queries): andreiaalexa/adaptive-rag-scifact-50

Citation

If you use SciFact, please cite the original dataset:

@inproceedings{wadden-etal-2020-fact,
    title = "Fact or Fiction: Verifying Scientific Claims",
    author = "Wadden, David and Lin, Shanchuan and Lo, Kyle and Wang, Lucy Lu and van Zuylen, Madeleine and Cohan, Arman and Hajishirzi, Hannaneh",
    booktitle = "Proceedings of EMNLP 2020",
    year = "2020"
}