English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

We perform knowledge distillation (KD) benchmark from task-specific BERT-baseteacher models to various student models: BiLSTM, CNN, BERT-Tiny, BERT-Mini,and BERT-Small. Our experiment involves 12 datasets grouped in two tasks: textclassification and sequence labeling in the Indonesian language. We alsocompare various aspects of distillations including the usage of word embeddingsand unlabeled data augmentation. Our experiments show that, despite the risingpopularity of Transformer-based models, using BiLSTM and CNN student modelsprovide the best trade-off between performance and computational resource (CPU,RAM, and storage) compared to pruned BERT models. We further propose some quickwins on performing KD to produce small NLP models via efficient KD trainingmechanisms involving simple choices of loss functions, word embeddings, andunlabeled data preparation.

Which Student is Best? A Comprehensive Knowledge Distillation Exam for Task-Specific BERT Models