This repository contains the publicly released dataset, code, and models for the Explanations for CommonsenseQA paper presented at ACL-IJCNLP 2021. Directories data and code inside the root folder contain dataset and code, respectively. The same data and code are also made available through our AIHN collaboration partner institute IIT Delhi. You can download the full paper from here.

Note that these annotations are provided for the questions of the CommonsenseQA data (https://www.tau-nlp.org/commonsenseqa): arXiv:1811.00937 [cs.CL] (or arXiv:1811.00937v2 [cs.CL] for this version).

Citations

Please consider citing this paper as follows:

@inproceedings{aggarwaletal2021ecqa,
  title={{E}xplanations for {C}ommonsense{QA}: {N}ew {D}ataset and {M}odels},
  author={Shourya Aggarwal and Divyanshu Mandowara and Vishwajeet Agrawal and Dinesh Khandelwal and Parag Singla and Dinesh Garg},
  booktitle="Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)}",
  Pages = 3050–3065,
  year = "2021",
  publisher = "Association for Computational Linguistics"
}