# The RepEval 2017 Shared Task: Multi-Genre Natural Language Inference with Sentence Representations

Source: [The RepEval 2017 Shared Task: Multi-Genre Natural Language Inference with Sentence Representations](<https://www.sota2.com/research/paper/68fa63821e209131ce9ff0a5>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-25T20:42:16.909Z

Authors: Nikita Nangia, Adina Williams, Angeliki Lazaridou, Samuel R. Bowman

Published: 2017-07-25T19:27:05.000Z

## Abstract

This paper presents the results of the RepEval 2017 Shared Task, which evaluated neural network sentence representation learning models on the Multi-Genre Natural Language Inference corpus (MultiNLI) recently introduced by Williams et al. (2017). All of the five participating teams beat the bidirectional LSTM (BiLSTM) and continuous bag of words baselines reported in Williams et al.. The best single model used stacked BiLSTMs with residual connections to extract sentence features and reached 74.5% accuracy on the genre-matched test set. Surprisingly, the results of the competition were fairly consistent across the genre-matched and genre-mismatched test sets, and across subsets of the test data representing a variety of linguistic phenomena, suggesting that all of the submitted systems learned reasonably domain-independent representations for sentence meaning.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Natural Language Inference on MultiNLI matched (in-domain) RepEval 2017 (test)](<https://www.sota2.com/research/sota/natural-language-inference-on-multinli-matched-in-domain-repeval-2017-test.md>): Natural Language Inference; MultiNLI matched (in-domain) RepEval 2017 (test); Accuracy: 74.5.

- [Natural Language Inference on MultiNLI mismatched (cross-domain) RepEval 2017 (test)](<https://www.sota2.com/research/sota/natural-language-inference-on-multinli-mismatched-cross-domain-repeval-2017-test.md>): Natural Language Inference; MultiNLI mismatched (cross-domain) RepEval 2017 (test); Accuracy: 73.5.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/1707.08172>)

[PDF](<https://arxiv.org/pdf/1707.08172.pdf>)