We introduce a new large-scale NLI benchmark dataset, collected via an iterative, adversarial human-and-model-in-the-loop procedure. We show that training models on this new dataset leads to state-of-the-art performance on a variety of popular NLI benchmarks, while posing a more difficult challenge with its new test set. Our analysis sheds light on the shortcomings of current state-of-the-art models, and shows that non-expert annotators are successful at finding their weaknesses. The data collection method can be applied in a never-ending learning scenario, becoming a moving target for NLU, rather than a static benchmark that will quickly saturate.
@article{arxiv.1910.14599,
title = {Adversarial NLI: A New Benchmark for Natural Language Understanding},
author = {Yixin Nie and Adina Williams and Emily Dinan and Mohit Bansal and Jason Weston and Douwe Kiela},
journal= {arXiv preprint arXiv:1910.14599},
year = {2020}
}