On the Use of Context for Predicting Citation Worthiness of Sentences in Scholarly Articles

Abstract

In this paper, we study the importance of context in predicting the citation worthiness of sentences in scholarly articles. We formulate this problem as a sequence labeling task solved using a hierarchical BiLSTM model. We contribute a new benchmark dataset containing over two million sentences and their corresponding labels. We preserve the sentence order in this dataset and perform document-level train/test splits, which importantly allows incorporating contextual information in the modeling process. We evaluate the proposed approach on three benchmark datasets. Our results quantify the benefits of using context and contextual embeddings for citation worthiness. Lastly, through error analysis, we provide insights into cases where context plays an essential role in predicting citation worthiness. Conclusion 0.

CiteCaseLAW: Citation Worthiness Detection in Caselaw for Legal Assistive Writing

“…Farber et al [3] and Bonab et al [4] utilized convolutional recurrent neural networks on diverse datasets. Context-aware citation detection was introduced by Gosangi et al [5] with the ACL-cite dataset, integrating BiLSTMs and transformer-based embeddings. Wright et al [6] delved into citation worthiness extensively, incorporating domain adaptation and transfer learning techniques.…”

“…We experimented with different models trained on our dataset to establish the baselines for the task of citation-worthiness detection (RQ2). For this assessment, we used our subset with 1M entries 5 . The split contains sentences sampled over all jurisdictions.…”

“…Thanking Rajiv Ratn Shah, who is partly supported by the Infosys Center for AI, the Center for Design and New Media, and the Center of Excellence in Healthcare at IIIT Delhi. We also want to thank Dr. Debanjan Mahata for motivating us and providing insights on the task [5].…”