TaCL: Improving BERT Pre-training with Token-aware Contrastive Learning

Overview

TaCL: Improving BERT Pre-training with Token-aware Contrastive Learning

Authors: Yixuan Su, Fangyu Liu, Zaiqiao Meng, Lei Shu, Ehsan Shareghi, and Nigel Collier

Code of our paper: TaCL: Improving BERT Pre-training with Token-aware Contrastive Learning

Introduction:

Masked language models (MLMs) such as BERT and RoBERTa have revolutionized the field of Natural Language Understanding in the past few years. However, existing pre-trained MLMs often output an anisotropic distribution of token representations that occupies a narrow subset of the entire representation space. Such token representations are not ideal, especially for tasks that demand discriminative semantic meanings of distinct tokens. In this work, we propose TaCL (Token-aware Contrastive Learning), a novel continual pre-training approach that encourages BERT to learn an isotropic and discriminative distribution of token representations. TaCL is fully unsupervised and requires no additional data. We extensively test our approach on a wide range of English and Chinese benchmarks. The results show that TaCL brings consistent and notable improvements over the original BERT model. Furthermore, we conduct detailed analysis to reveal the merits and inner-workings of our approach

Main Results:

We show the comparison between TaCL (base version) and the original BERT (base version).

(1) English benchmark results on SQuAD (Rajpurkar et al., 2018) (dev set) and GLUE (Wang et al., 2019) average score.

Model SQuAD 1.1 (EM/F1) SQuAD 2.0 (EM/F1) GLUE Average
BERT 80.8/88.5 73.4/76.8 79.6
TaCL 81.6/89.0 74.4/77.5 81.2

(2) Chinese benchmark results (test set F1) on four NER tasks (MSRA, OntoNotes, Resume, and Weibo) and three Chinese word segmentation (CWS) tasks (PKU, CityU, and AS).

Model MSRA OntoNotes Resume Weibo PKU CityU AS
BERT 94.95 80.14 95.53 68.20 96.50 97.60 96.50
TaCL 95.44 82.42 96.45 69.54 96.75 98.16 96.75

Huggingface Models:

Model Name Model Address
English (cambridgeltl/tacl-bert-base-uncased) link
Chinese (cambridgeltl/tacl-bert-base-chinese) link

Example Usage:

import torch
# initialize model
from transformers import AutoModel, AutoTokenizer
model_name = 'cambridgeltl/tacl-bert-base-uncased'
model = AutoModel.from_pretrained(model_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)
# create input ids
text = '[CLS] clbert is awesome. [SEP]'
tokenized_token_list = tokenizer.tokenize(text)
input_ids = torch.LongTensor(tokenizer.convert_tokens_to_ids(tokenized_token_list)).view(1, -1)
# compute hidden states
representation = model(input_ids).last_hidden_state # [1, seqlen, embed_dim]

Tutorial (in Chinese language) on how to use Chinese TaCL BERT to performance Name Entity Recognition and Chinese word segmentation:

Tutorial link

Tutorial on how to reproduce the results in our paper:

1. Environment Setup:

python version: 3.8
pip3 install -r requirements.txt

2. Train TaCL:

(1) Prepare pre-training data:

Please refer to details provided in ./pretraining_data directory.

(2) Train the model:

Please refer to details provided in ./pretraining directory.

3. Experiments on English Benchmarks:

Please refer to details provided in ./english_benchmark directory.

4. Experiments on Chinese Benchmarks:

(1) Chinese Benchmark Data Preparation:

chmod +x ./download_benchmark_data.sh
./download_benchmark_data.sh

(2) Fine-tuning and Inference:

Please refer to details provided in ./chinese_benchmark directory.

5. Replicate Our Analysis Results:

We provide all essential code to replicate the results (the images below) provided in our analysis section. The related codes and instructions are located in ./analysis directory. Have fun!

Citation:

If you find our paper and resources useful, please kindly cite our paper:

@misc{su2021tacl,
      title={TaCL: Improving BERT Pre-training with Token-aware Contrastive Learning}, 
      author={Yixuan Su and Fangyu Liu and Zaiqiao Meng and Lei Shu and Ehsan Shareghi and Nigel Collier},
      year={2021},
      eprint={2111.04198},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}

Contact

If you have any questions, feel free to contact me via ([email protected]).

Owner
Yixuan Su
I am a final-year PhD student at the University of Cambridge, supervised by Professor Nigel Collier.
Yixuan Su
PSML: A Multi-scale Time-series Dataset for Machine Learning in Decarbonized Energy Grids

PSML: A Multi-scale Time-series Dataset for Machine Learning in Decarbonized Energy Grids The electric grid is a key enabling infrastructure for the a

Texas A&M Engineering Research 19 Jan 07, 2023
Official implementation of "Not only Look, but also Listen: Learning Multimodal Violence Detection under Weak Supervision" ECCV2020

XDVioDet Official implementation of "Not only Look, but also Listen: Learning Multimodal Violence Detection under Weak Supervision" ECCV2020. The proj

peng 64 Dec 12, 2022
The final project of "Applying AI to 3D Medical Imaging Data" from "AI for Healthcare" nanodegree - Udacity.

Quantifying Hippocampus Volume for Alzheimer's Progression Background Alzheimer's disease (AD) is a progressive neurodegenerative disorder that result

Omar Laham 1 Jan 14, 2022
🐤 Nix-TTS: An Incredibly Lightweight End-to-End Text-to-Speech Model via Non End-to-End Distillation

🐤 Nix-TTS An Incredibly Lightweight End-to-End Text-to-Speech Model via Non End-to-End Distillation Rendi Chevi, Radityo Eko Prasojo, Alham Fikri Aji

Rendi Chevi 156 Jan 09, 2023
PointRCNN: 3D Object Proposal Generation and Detection from Point Cloud, CVPR 2019.

PointRCNN PointRCNN: 3D Object Proposal Generation and Detection from Point Cloud Code release for the paper PointRCNN:3D Object Proposal Generation a

Shaoshuai Shi 1.5k Dec 27, 2022
Facestar dataset. High quality audio-visual recordings of human conversational speech.

Facestar Dataset Description Existing audio-visual datasets for human speech are either captured in a clean, controlled environment but contain only a

Meta Research 87 Dec 21, 2022
The official implementation of Variable-Length Piano Infilling (VLI).

Variable-Length-Piano-Infilling The official implementation of Variable-Length Piano Infilling (VLI). (paper: Variable-Length Music Score Infilling vi

29 Sep 01, 2022
Codes and models of NeurIPS2021 paper - DominoSearch: Find layer-wise fine-grained N:M sparse schemes from dense neural networks

DominoSearch This is repository for codes and models of NeurIPS2021 paper - DominoSearch: Find layer-wise fine-grained N:M sparse schemes from dense n

11 Sep 10, 2022
This repository contains the code for the CVPR 2020 paper "Differentiable Volumetric Rendering: Learning Implicit 3D Representations without 3D Supervision"

Differentiable Volumetric Rendering Paper | Supplementary | Spotlight Video | Blog Entry | Presentation | Interactive Slides | Project Page This repos

697 Jan 06, 2023
Official implementation of NeuralFusion: Online Depth Map Fusion in Latent Space

NeuralFusion This is the official implementation of NeuralFusion: Online Depth Map Fusion in Latent Space. We provide code to train the proposed pipel

53 Jan 01, 2023
Scalable Attentive Sentence-Pair Modeling via Distilled Sentence Embedding (AAAI 2020) - PyTorch Implementation

Scalable Attentive Sentence-Pair Modeling via Distilled Sentence Embedding PyTorch implementation for the Scalable Attentive Sentence-Pair Modeling vi

Microsoft 25 Dec 02, 2022
[NeurIPS 2021] Better Safe Than Sorry: Preventing Delusive Adversaries with Adversarial Training

Better Safe Than Sorry: Preventing Delusive Adversaries with Adversarial Training Code for NeurIPS 2021 paper "Better Safe Than Sorry: Preventing Delu

Lue Tao 29 Sep 20, 2022
scAR (single-cell Ambient Remover) is a package for data denoising in single-cell omics.

scAR scAR (single cell Ambient Remover) is a package for denoising multiple single cell omics data. It can be used for multiple tasks, such as, sgRNA

19 Nov 28, 2022
TCNN Temporal convolutional neural network for real-time speech enhancement in the time domain

TCNN Pandey A, Wang D L. TCNN: Temporal convolutional neural network for real-time speech enhancement in the time domain[C]//ICASSP 2019-2019 IEEE Int

凌逆战 16 Dec 30, 2022
Activity image-based video retrieval

Cross-modal-retrieval Our approach is focus on Activity Image-to-Video Retrieval (AIVR) task. The compared methods are state-of-the-art single modalit

BCMI 75 Oct 21, 2021
Motion Reconstruction Code and Data for Skills from Videos (SFV)

Motion Reconstruction Code and Data for Skills from Videos (SFV) This repo contains the data and the code for motion reconstruction component of the S

268 Dec 01, 2022
QA-GNN: Question Answering using Language Models and Knowledge Graphs

QA-GNN: Question Answering using Language Models and Knowledge Graphs This repo provides the source code & data of our paper: QA-GNN: Reasoning with L

Michihiro Yasunaga 434 Jan 04, 2023
Code for Fully Context-Aware Image Inpainting with a Learned Semantic Pyramid

SPN: Fully Context-Aware Image Inpainting with a Learned Semantic Pyramid Code for Fully Context-Aware Image Inpainting with a Learned Semantic Pyrami

12 Jun 27, 2022
Pytorch Lightning Implementation of SC-Depth Methods.

SC_Depth_pl: This is a pytorch lightning implementation of SC-Depth (V1, V2) for self-supervised learning of monocular depth from video. In the V1 (IJ

JiaWang Bian 216 Dec 30, 2022
An implementation of Geoffrey Hinton's paper "How to represent part-whole hierarchies in a neural network" in Pytorch.

GLOM An implementation of Geoffrey Hinton's paper "How to represent part-whole hierarchies in a neural network" for MNIST Dataset. To understand this

50 Oct 19, 2022