Multi-Vector Models with Textual Guidance for Fine-Grained Scientific Document Similarity

Anonymous

Paper Link: https://openreview.net/forum?id=yoZhDj8j8dK

Paper Type: Long paper (up to eight pages of content + unlimited references and appendices)

Abstract: We present a new scientific document similarity model based on matching fine-grained aspects of texts. To train our model, we exploit a naturally-occurring source of supervision: sentences in the full-text of papers that cite multiple papers together (co-citations). Such co-citations not only reflect close paper relatedness, but also provide textual descriptions of how the co-cited papers are related. This novel form of textual supervision is used for learning to match aspects across papers. We develop multi-vector representations where vectors correspond to sentence-level aspects of documents, and present two methods for aspect matching: (1) A fast method that only matches single aspects, and (2) a method that makes sparse multiple matches with an Optimal Transport mechanism that computes an Earth Mover's Distance between aspects. Our approach improves performance on document similarity tasks in four datasets. Further, our fast single-match method achieves competitive results, paving the way for applying fine-grained similarity to large scientific corpora.

Presentation Mode: This paper will be presented in person in Seattle

Copyright Consent Signature (type Name Or NA If Not Transferrable): Sheshera Mysore

Copyright Consent Name And Address: Sheshera Mysore, Manning College of Information and Computer Sciences, 140 Governors Dr., Amherst, MA 01003