Skip to content

Thorsten-Trinkaus/NERC

Repository files navigation

GitHub License GitHub last commit

NERC of Different Granularities

Course project for Formale Semantik (University of Heidelberg). We investigate Named Entity Recognition & Classification (NERC) under increasing label granularity (from coarse-grained up to ultra-fine entity typing).

Datasets

OntoNotes: The 90% Solution (Hovy et al., NAACL 2006)

Fine-grained entity recognition (FIGER) (Ling and Weld, AAAI 2012)

Ultra-Fine Entity Typing (Choi et al., ACL 2018)


Status: Work in progress (this repo will evolve as experiments and structure solidify!).

About

Project to investigate how pre-trained language models handle NER/NERC as label granularity increases, comparing fine-tuning, probing, and NLI-based approaches.

Topics

Resources

License

Code of conduct

Stars

Watchers

Forks

Contributors