Shufan Wang

TL;DR: The STORIUM dataset as discussed by the authors contains 6K lengthy stories with fine-grained natural language annotations (e.g., character goals and attributes) interspersed throughout each narrative, forming a robust source for guiding models.

TL;DR: This paper proposed a contrastive fine-tuning objective that enables BERT to produce more powerful phrase embeddings, which can be integrated with a simple autoencoder to build a phrase-based neural topic model that interprets topics as mixtures of words and phrases.

TL;DR: In this paper, the academic structure of Trinity College in Hartford, CT using the major and minor patterns between graduating students to build a temporal multiplex network describing the interactions between different departments.

TL;DR: This paper uses the major and minor patterns between graduating students to build a temporal multiplex network describing the interactions between different departments, and identifies the evolving community structures that organize departments’ interactions, as well as quantify the interdisciplinary centrality of each department.