English

CRoW: Benchmarking Commonsense Reasoning in Real-World Tasks

Computation and Language 2023-10-25 v1 Artificial Intelligence

Abstract

Recent efforts in natural language processing (NLP) commonsense reasoning research have yielded a considerable number of new datasets and benchmarks. However, most of these datasets formulate commonsense reasoning challenges in artificial scenarios that are not reflective of the tasks which real-world NLP systems are designed to solve. In this work, we present CRoW, a manually-curated, multi-task benchmark that evaluates the ability of models to apply commonsense reasoning in the context of six real-world NLP tasks. CRoW is constructed using a multi-stage data collection pipeline that rewrites examples from existing datasets using commonsense-violating perturbations. We use CRoW to study how NLP systems perform across different dimensions of commonsense knowledge, such as physical, temporal, and social reasoning. We find a significant performance gap when NLP systems are evaluated on CRoW compared to humans, showcasing that commonsense reasoning is far from being solved in real-world task settings. We make our dataset and leaderboard available to the research community at https://github.com/mismayil/crow.

Keywords

Cite

@article{arxiv.2310.15239,
  title  = {CRoW: Benchmarking Commonsense Reasoning in Real-World Tasks},
  author = {Mete Ismayilzada and Debjit Paul and Syrielle Montariol and Mor Geva and Antoine Bosselut},
  journal= {arXiv preprint arXiv:2310.15239},
  year   = {2023}
}

Comments

37 pages, camera-ready for EMNLP 2023

R2 v1 2026-06-28T12:59:25.662Z