Unsupervised Pronoun Resolution via Masked Noun-Phrase Prediction
This addresses pronoun resolution for natural language processing applications, representing a strong specific gain rather than a foundational advancement.
The paper tackles pronoun resolution in an unsupervised setting by proposing Masked Noun-Phrase Prediction (MNPP), achieving large-margin improvements over previous unsupervised methods on all datasets and outperforming RoBERTa-large in few-shot settings with higher AUC scores.
In this work, we propose Masked Noun-Phrase Prediction (MNPP), a pre-training strategy to tackle pronoun resolution in a fully unsupervised setting. Firstly, We evaluate our pre-trained model on various pronoun resolution datasets without any finetuning. Our method outperforms all previous unsupervised methods on all datasets by large margins. Secondly, we proceed to a few-shot setting where we finetune our pre-trained model on WinoGrande-S and XS separately. Our method outperforms RoBERTa-large baseline with large margins, meanwhile, achieving a higher AUC score after further finetuning on the remaining three official splits of WinoGrande.