We present the task of modeling information propagation in literature, in which we seek to identify pieces of information passing from character A to character B to character C, only given a description of their activity in text. We describe a new pipeline for measuring information propagation in this domain and publish a new dataset for speaker attribution, enabling the evaluation of an important component of this pipeline on a wider range of literary texts than previously studied. Using this pipeline, we analyze the dynamics of information propagation in over 5,000 works of fiction, finding that information flows through characters that fill structural holes connecting different communities, and that characters who are women are depicted as filling this role much more frequently than characters who are men.

We present in this work a method for incorporating global context in long documents when making local decisions in sequence labeling problems like NER. Inspired by work in featurized log-linear models (Chieu and Ng, 2002; Sutton and McCallum, 2004), our model learns to attend to multiple mentions of the same word type in generating a representation for each token in context, extending that work to learning representations that can be incorporated into modern neural models. Attending to broader context at test time provides complementary information to pretraining (Gururangan et al., 2020), yields strong gains over equivalently parameterized models lacking such context, and performs best at recognizing entities with high TF-IDF scores (i.e., those that are important within a document).

In this work we present a new dataset of literary events—events that are depicted as taking place within the imagined space of a novel. While previous work has focused on event detection in the domain of contemporary news, literature poses a number of complications for existing systems, including complex narration, the depiction of a broad array of mental states, and a strong emphasis on figurative language. We outline the annotation decisions of this new dataset and compare several models for predicting events; the best performing model, a bidirectional LSTM with BERT token representations, achieves an F1 score of 73.9. We then apply this model to a corpus of novels split across two dimensions—prestige and popularity—and demonstrate that there are statistically significant differences in the distribution of events for prestige.