TY - JOUR
AU - 
AB - A Broad-Coverage Challenge Corpus for Sentence Understanding through Inference 1 2 1,2,3 Adina Williams Nikita Nangia Samuel R. Bowman adinawilliams@nyu.edu nikitanangia@nyu.edu bowman@nyu.edu 1 2 3 Department of Linguistics Center for Data Science Department of Computer Science New York University New York University New York University Abstract which current models extract reasonable represen- tations of language meaning in these settings. This paper introduces the Multi-Genre Natu- The task of natural language inference (NLI) ral Language Inference (MultiNLI) corpus, a is well positioned to serve as a benchmark task dataset designed for use in the development and evaluation of machine learning models for for research on NLU. In this task, also known sentence understanding. At 433k examples, as recognizing textual entailment (Cooper et al., this resource is one of the largest corpora avail- 1996; Fyodorov et al., 2000; Condoravdi et al., able for natural language inference (a.k.a. rec- 2003; Bos and Markert, 2005; Dagan et al., 2006; ognizing textual entailment), improving upon MacCartney and Manning, 2009), a model is pre- available resources in both its coverage and sented with a pair of sentences—like one of those difficulty. MultiNLI accomplishes this by of- in Figure 1—and asked to judge the relationship fering 
TI - A Broad-Coverage Challenge Corpus for Sentence Understanding through Inference
JF - Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers)
DO - 10.18653/v1/n18-1101
DA - 2018-01-01
UR - https://www.deepdyve.com/lp/unpaywall/a-broad-coverage-challenge-corpus-for-sentence-understanding-through-L1CV0iyY2o
DP - DeepDyve
ER -