1 paper
Benjamin Townsend, Eamon Ito-Fisher, Lily Zhang +1
Typically, information extraction (IE) requires a pipeline approach: first, a sequence labeling model is trained on manually annotated documents to extract relevant spans; then, wh…