论文信息 - TTTTTackling WinoGrande Schemas

TTTTTackling WinoGrande Schemas

We applied the T5 sequence-to-sequence model to tackle the AI2 WinoGrande Challenge by decomposing each example into two input text strings, each containing a hypothesis, and using the probabilities assigned to the "entailment" token as a score of the hypothesis. Our first (and only) submission to the official leaderboard yielded 0.7673 AUC on March 13, 2020, which is the best known result at this time and beats the previous state of the art by over five points.

[1] Samuel R. Bowman,et al. A Broad-Coverage Challenge Corpus for Sentence Understanding through Inference , 2017, NAACL.

[2] Terry Winograd,et al. Understanding natural language , 1974 .

[3] Yejin Choi,et al. WINOGRANDE: An Adversarial Winograd Schema Challenge at Scale , 2020, AAAI.

[4] Omer Levy,et al. BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension , 2019, ACL.

[5] Yiming Yang,et al. XLNet: Generalized Autoregressive Pretraining for Language Understanding , 2019, NeurIPS.

[6] Ming-Wei Chang,et al. BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding , 2019, NAACL.

[7] Jimmy J. Lin,et al. Document Ranking with a Pretrained Sequence-to-Sequence Model , 2020, FINDINGS.

[8] Omer Levy,et al. RoBERTa: A Robustly Optimized BERT Pretraining Approach , 2019, ArXiv.

[9] Colin Raffel,et al. Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer , 2019, J. Mach. Learn. Res..