Skip to content

Implementation of a "End-to-end Neural Coreference Resolution" model. *Note* end goal is to compare LLM with neural models.

License

Notifications You must be signed in to change notification settings

weezymatt/e2e-corefer

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

34 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Coreference Resolution with CoNLL-U or CorefUD

A fully functional Coreference Resolution project.

What is coreference resolution?

Coreference is a kind of reference resolution that aims to resolve the relationship between which mentions are linked (or corefer) to the same entity. Applications include a wide variety of tasks, such as knowledge extraction, question answering, and dialog systems.

Installation

To explore coreference Use the package manager to install pip to install the CoNLL-U Parser.

pip install conllu

Or, if you fancy conda:

conda install -c conda-forge conllu

For more information, please click here.

CoNLL-U Format

The CoNLL-U format is used by the Universal Dependencies project and serves as a standardized approach to portray annotated linguistic data, generally related to tasks in Natural Language Processing. Our primary purpose using the format is to provide

References

We've been inspired by work attempting to understand the coreference resolution system. Additionally, recent work has produced an initiative to collect coreference corpora in various languages and harmonize them (CoNLL-U).

[1]: https://aclanthology.org/2023.crac-sharedtask.1.pdf

[2]: https://aclanthology.org/2022.lrec-1.520.pdf

About

Implementation of a "End-to-end Neural Coreference Resolution" model. *Note* end goal is to compare LLM with neural models.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published