ZeroSCROLLS: Zero-Shot CompaRison Over Long Language Sequences
​
What is Zero-SCROLLS?
ZeroSCROLLS is a suite of datasets that require synthesizing information over long texts. The benchmark includes ten natural language tasks across multiple domains, including summarization, question answering, aggregated sentiment classification and information reordering.
​
​
​
​
Citing ZeroSCROLLS
Please use the following bibliography to cite ZeroSCROLLS:
@misc{shaham2023zeroscrolls,
title={ZeroSCROLLS: A Zero-Shot Benchmark for Long Text Understanding},
author={Uri Shaham and Maor Ivgi and Avia Efrat and Jonathan Berant and Omer Levy},
year={2023},
eprint={2305.14196},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
When citing ZeroSCROLLS, please make sure to cite all of the original dataset papers. [bibtex]
​
Contact Us
