SWAG Table

These are the SWAG test results after fine-tuning: $$BERT_{LARGE}$$ outperforms the authors’ baseline ESIM+ELMo system by +27.1% and OpenAI GPT by 8.3%.

Ohio State University, Columbus

Google

The SWAG dataset contains 113k sentence-pair completion examples that evaluate grounded common-sense inference.

Situations With Adversarial Generations (SWAG)


Devlin, J., Chang, M. W., Lee, K., & Toutanova, K. (2018). Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805.

https://paperswithcode.com/method/bert

Learn Before

Related

Learn After