Search

The article addresses document-level event extraction challenges, proposing a novel framework with a token-event-role data structure and a multi-channel prediction module. Topics include the significance of document-level event extraction, challenges in multi-event detection and argument identification, and previous methods' limitations.

The article focuses on rHDP, a hierarchical topic model for multi-domain corpora, addressing challenges in identifying inter-topic correlations and aspect-sharing patterns. Topics include the enhancement of nested Chinese restaurant processes for aspect-sharing pattern extraction, the development of a hierarchical Gibbs sampling scheme for model inference, and the validation of the model's efficacy through experiments on public and financial document datasets.