# ERNIE-Tiny : A Progressive Distillation Framework for Pretrained Transformer Compression

arXiv: https://arxiv.org/abs/2106.02241 · HTML: https://mlsanity.com/paper/2106.02241/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Builds on 40 of its 53 references that are in the map.
- First posted on arXiv 4 Jun 2021.

## Builds on (most cited first, 30 of 40)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Explaining and Harnessing Adversarial Examples](https://mlsanity.com/paper/1412.6572/) (2014)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension](https://mlsanity.com/paper/1910.13461/) (2019)
- [Layer Normalization](https://mlsanity.com/paper/1607.06450/) (2016)
- [GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding](https://mlsanity.com/paper/1804.07461/) (2018)
- [Unsupervised Cross-lingual Representation Learning at Scale](https://mlsanity.com/paper/1911.02116/) (2019)
- [Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer](https://mlsanity.com/paper/1701.06538/) (2017)
- [Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity](https://mlsanity.com/paper/2101.03961/) (2021)
- [XLNet: Generalized Autoregressive Pretraining for Language Understanding](https://mlsanity.com/paper/1906.08237/) (2019)
- [ALBERT: A Lite BERT for Self-supervised Learning of Language Representations](https://mlsanity.com/paper/1909.11942/) (2019)
- [Learning both Weights and Connections for Efficient Neural Networks](https://mlsanity.com/paper/1506.02626/) (2015)
- [Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism](https://mlsanity.com/paper/1909.08053/) (2019)
- [A large annotated corpus for learning natural language inference](https://mlsanity.com/paper/1508.05326/) (2015)
- [MiniLM: Deep Self-Attention Distillation for Task-Agnostic Compression of Pre-Trained Transformers](https://mlsanity.com/paper/2002.10957/) (2020)
- [Energy and Policy Considerations for Deep Learning in NLP](https://mlsanity.com/paper/1906.02243/) (2019)
- [TinyBERT: Distilling BERT for Natural Language Understanding](https://mlsanity.com/paper/1909.10351/) (2019)
- [Cross-lingual Language Model Pretraining](https://mlsanity.com/paper/1901.07291/) (2019)
- [Temporal Ensembling for Semi-Supervised Learning](https://mlsanity.com/paper/1610.02242/) (2016)
- [Aligning Books and Movies: Towards Story-like Visual Explanations by Watching Movies and Reading Books](https://mlsanity.com/paper/1506.06724/) (2015)
- [Virtual Adversarial Training: A Regularization Method for Supervised and Semi-Supervised Learning](https://mlsanity.com/paper/1704.03976/) (2017)
- [XNLI: Evaluating Cross-lingual Sentence Representations](https://mlsanity.com/paper/1809.05053/) (2018)
- [SpanBERT: Improving Pre-training by Representing and Predicting Spans](https://mlsanity.com/paper/1907.10529/) (2019)
- [Patient Knowledge Distillation for BERT Model Compression](https://mlsanity.com/paper/1908.09355/) (2019)
- [MobileBERT: a Compact Task-Agnostic BERT for Resource-Limited Devices](https://mlsanity.com/paper/2004.02984/) (2020)

## Built on by (most cited first, 8 of 8)

- [ERNIE 3.0 Titan: Exploring Larger-scale Knowledge Enhanced Pre-training for Language Understanding and Generation](https://mlsanity.com/paper/2112.12731/) (2021)
- [A Closer Look at Self-Supervised Lightweight Vision Transformers](https://mlsanity.com/paper/2205.14443/) (2022)
- [PROD: Progressive Distillation for Dense Retrieval](https://mlsanity.com/paper/2209.13335/) (2022)
- [NutePrune: Efficient Progressive Pruning with Numerous Teachers for Large Language Models](https://mlsanity.com/paper/2402.09773/) (2024)
- [ERNIE 3.0 Tiny: Frustratingly Simple Method to Improve Task-Agnostic Distillation Generalization](https://mlsanity.com/paper/2301.03416/) (2023)
- [GOVERN: Gradient Orientation Vote Ensemble for Multi-Teacher Reinforced Distillation](https://mlsanity.com/paper/2405.03764/) (2024)
- [GenDistiller: Distilling Pre-trained Language Models based on Generative Models](https://mlsanity.com/paper/2310.13418/) (2023)
- [GenDistiller: Distilling Pre-trained Language Models based on an Autoregressive Generative Model](https://mlsanity.com/paper/2406.09444/) (2024)

## Abstract (arXiv)

Pretrained language models (PLMs) such as BERT adopt a training paradigm which first pretrain the model in general data and then finetune the model on task-specific data, and have recently achieved great success. However, PLMs are notorious for their enormous parameters and hard to be deployed on real-life applications. Knowledge distillation has been prevailing to address this problem by transferring knowledge from a large teacher to a much smaller student over a set of data. We argue that the selection of thee three key components, namely teacher, training data, and learning objective, is crucial to the effectiveness of distillation. We, therefore, propose a four-stage progressive distillation framework ERNIE-Tiny to compress PLM, which varies the three components gradually from general level to task-specific level. Specifically, the first stage, General Distillation, performs distillation with guidance from pretrained teacher, gerenal data and latent distillation loss. Then, General-Enhanced Distillation changes teacher model from pretrained teacher to finetuned teacher. After that, Task-Adaptive Distillation shifts training data from general data to task-specific data. In the end, Task-Specific Distillation, adds two additional losses, namely Soft-Label and Hard-Label loss onto the last stage. Empirical results demonstrate the effectiveness of our framework and generalization gain brought by ERNIE-Tiny.In particular, experiments show that a 4-layer ERNIE-Tiny maintains over 98.0%performance of its 12-layer teacher BERT base on GLUE benchmark, surpassing state-of-the-art (SOTA) by 1.0% GLUE score with the same amount of parameters. Moreover, ERNIE-Tiny achieves a new compression SOTA on five Chinese NLP tasks, outperforming BERT base by 0.4% accuracy with 7.5x fewer parameters and9.4x faster inference speed.