Domain Knowledge Transferring for Pre-trained Language Model via Calibrated Activation Boundary Distillation

1 code implementation • ACL 2022 • Dongha Choi, HongSeok Choi, Hyunju Lee

In this study, we propose a domain knowledge transferring (DoKTra) framework for PLMs without additional in-domain pretraining.

Knowledge Distillation Language Modelling

Early Stopping Based on Unlabeled Samples in Text Classification

1 code implementation • ACL 2022 • HongSeok Choi, Dongha Choi, Hyunju Lee

The proposed method is advantageous because it does not require a separate validation set and provides a better stopping point by using a large unlabeled set.

text-classification Text Classification

Extracting Chemical-Protein Interactions via Calibrated Deep Neural Network and Self-training

no code implementations • Findings of the Association for Computational Linguistics 2020 • Dongha Choi, Hyunju Lee

To estimate the data uncertainty and improve the reliability, "calibration" techniques have been applied to deep learning models.