Skip to content
/ CENSOR Public

[ACL'24] Code for "Improving the Robustness of Distantly-Supervised Named Entity Recognition via Uncertainty-Aware Teacher Learning and Student-Student Collaborative Learning"

Notifications You must be signed in to change notification settings

S1s-Z/CENSOR

Repository files navigation

CENSOR

Hi, this is the code of our paper "Improving the Robustness of Distantly-Supervised Named Entity Recognition via Uncertainty-Aware Teacher Learning and Student-Student Collaborative Learning" accepted by ACL 2024 Findings. Our paper is available here.

News:

Accepted by ACL 2024 Findings. 2024.05

Code released at Github. 2024.09

Preparation

Use requirements.txt to get the right environments

Reproduce results

We provide three bash scripts run_conll03.sh, run_ontonotes5.sh, run_webpage.sh, run_wikigold.sh, run_twitter.sh for running the model on the datasets.

you can run the code like:

 sh <run_dataset>.sh <GPU ID> <DATASET NAME>

e.g.

sh run_conll03.sh 0,1 conll03

We got our results in single A40.

About

[ACL'24] Code for "Improving the Robustness of Distantly-Supervised Named Entity Recognition via Uncertainty-Aware Teacher Learning and Student-Student Collaborative Learning"

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published