Coreferential Reasoning Learning for Language Representation
This addresses the need for better coreference resolution in NLP models, which is crucial for tasks like question answering and summarization, representing an incremental improvement over existing methods.
The paper tackles the problem that most language representation models cannot explicitly handle coreference, which is essential for coherent discourse understanding, by introducing CorefBERT, a model that captures coreferential relations in context, achieving significant improvements on various downstream NLP tasks requiring coreferential reasoning while maintaining comparable performance on other tasks.
Language representation models such as BERT could effectively capture contextual semantic information from plain text, and have been proved to achieve promising results in lots of downstream NLP tasks with appropriate fine-tuning. However, most existing language representation models cannot explicitly handle coreference, which is essential to the coherent understanding of the whole discourse. To address this issue, we present CorefBERT, a novel language representation model that can capture the coreferential relations in context. The experimental results show that, compared with existing baseline models, CorefBERT can achieve significant improvements consistently on various downstream NLP tasks that require coreferential reasoning, while maintaining comparable performance to previous models on other common NLP tasks. The source code and experiment details of this paper can be obtained from https://github.com/thunlp/CorefBERT.