The implementation our EMNLP 2021 paper "Enhanced Language Representation with Label Knowledge for Span Extraction".

See below for an overview of the model architecture:

• python 3.6
• pytorch 1.8
• transformers 3.9.2
• prefetch_generator
• tokenizers

Note: The thunlp has updated the repo HMEAE recently, which causing the mismatch of data. Make sure you use the earlier version for ED task.

The details of hyper-parameter setting is listed in our paper - A.3.

• For NER task, run run_ner.py.
• For Event Detection task, run run_trigger_extraction.py

• Nested: set exist_nested=True.
• Flat: set span_decode_strategy=v5.