Hypothesis Only Baselines in Natural Language Inference

Adam Poliak, Jason Naradowsky, Aparajita Haldar, Rachel Rudinger, Benjamin Van Durme


Abstract
We propose a hypothesis only baseline for diagnosing Natural Language Inference (NLI). Especially when an NLI dataset assumes inference is occurring based purely on the relationship between a context and a hypothesis, it follows that assessing entailment relations while ignoring the provided context is a degenerate solution. Yet, through experiments on 10 distinct NLI datasets, we find that this approach, which we refer to as a hypothesis-only model, is able to significantly outperform a majority-class baseline across a number of NLI datasets. Our analysis suggests that statistical irregularities may allow a model to perform NLI in some datasets beyond what should be achievable without access to the context.
Anthology ID:
S18-2023
Volume:
Proceedings of the Seventh Joint Conference on Lexical and Computational Semantics
Month:
June
Year:
2018
Address:
New Orleans, Louisiana
Editors:
Malvina Nissim, Jonathan Berant, Alessandro Lenci
Venue:
*SEM
SIGs:
SIGLEX | SIGSEM
Publisher:
Association for Computational Linguistics
Note:
Pages:
180–191
Language:
URL:
https://aclanthology.org/S18-2023
DOI:
10.18653/v1/S18-2023
Bibkey:
Cite (ACL):
Adam Poliak, Jason Naradowsky, Aparajita Haldar, Rachel Rudinger, and Benjamin Van Durme. 2018. Hypothesis Only Baselines in Natural Language Inference. In Proceedings of the Seventh Joint Conference on Lexical and Computational Semantics, pages 180–191, New Orleans, Louisiana. Association for Computational Linguistics.
Cite (Informal):
Hypothesis Only Baselines in Natural Language Inference (Poliak et al., *SEM 2018)
Copy Citation:
PDF:
https://preview.aclanthology.org/dois-2013-emnlp/S18-2023.pdf
Code
 azpoliak/hypothesis-only-NLI
Data
Flickr30kMultiNLISICKSNLI