Repository for the paper titled: "When is BERT Multilingual? Isolating Crucial Ingredients for Cross-lingual Transfer"

Related tags

Deep Learningmultilingualtransformerszero-shotcross-lingualmultilingual-bert
Overview

When is BERT Multilingual? Isolating Crucial Ingredients for Cross-lingual Transfer

This repository contains code for our paper titled "When is BERT Multilingual? Isolating Crucial Ingredients for Cross-lingual Transfer". [arXiv]

Table of contents

  1. Paper in a nutshell
  2. Installation
  3. Data and models
  4. Repository usage
  5. Links to experiments and results
  6. Citation

Paper in a nutshell

While recent work on multilingual language models has demonstrated their capacity for cross-lingual zero-shot transfer on downstream tasks, there is a lack of consensus in the community as to what shared properties between languages enable such transfer. Analyses involving pairs of natural languages are often inconclusive and contradictory since languages simultaneously differ in many linguistic aspects. In this paper, we perform a large-scale empirical study to isolate the effects of various linguistic properties by measuring zero-shot transfer between four diverse natural languages and their counterparts constructed by modifying aspects such as the script, word order, and syntax. Among other things, our experiments show that the absence of sub-word overlap significantly affects zero-shot transfer when languages differ in their word order, and there is a strong correlation between transfer performance and word embedding alignment between languages (e.g., Spearman's R=0.94 on the task of NLI). Our results call for focus in multilingual models on explicitly improving word embedding alignment between languages rather than relying on its implicit emergence.

Installation instructions

  1. Step 1: Install from the conda .yml file.
conda env create -f installation/multilingual.yml
  1. Step 2: Install transformers in an editable way.
pip install -e transformers/
pip install -r transformers/examples/language-modeling/requirements.txt
pip install -r transformers/examples/token-classification/requirements.txt

Repository usage

For the commands we used to get the reported numbers in the paper, click here. This file contains common instructions used. This file can automatically generate commands for your use case.

Bilingual pre-training

  1. For bilingual pre-training on original and derived language pairs, use the flag --invert_word_order for the Inversion transformation, --permute_words for Permutation and --one_to_one_mapping for Transliteration. Example command for bilingual pre-training for English with Inversion transformation to create the derived language pair.
nohup  python transformers/examples/xla_spawn.py --num_cores 8 transformers/examples/language-modeling/run_mlm_synthetic.py --warmup_steps 10000 --learning_rate 1e-4 --save_steps -1 --max_seq_length 512 --logging_steps 50 --overwrite_output_dir --model_type roberta --config_name config/en/roberta_8/config.json --tokenizer_name config/en/roberta_8/ --do_train --do_eval --max_steps 500000 --per_device_train_batch_size 16 --per_device_eval_batch_size 16 --train_file ../../bucket/pretrain_data/en/train.txt --validation_file ../../bucket/pretrain_data/en/valid.txt --output_dir ../../bucket/model_outputs/en/inverted_order_500K/mlm --run_name inverted_en_500K_mlm --invert_word_order --word_modification add &
  1. For Syntax transformations, the train file used in the following command ([email protected][email protected]) means that it is the concatenation of French corpus with French modified to English verb and noun order ([email protected][email protected]).
nohup python transformers/examples/xla_spawn.py --num_cores 8 transformers/examples/language-modeling/run_mlm_synthetic.py --warmup_steps 10000 --learning_rate 1e-4 --save_steps -1 --max_seq_length 512 --logging_steps 50 --overwrite_output_dir --model_type roberta --config_name config/fr/roberta_8/config.json --tokenizer_name config/fr/roberta_8/ --do_train --do_eval --max_steps 500000 --per_device_train_batch_size 16 --per_device_eval_batch_size 16 --train_file ../../bucket/pretrain_data/fr/synthetic/[email protected][email protected] --validation_file ../../bucket/pretrain_data/fr/synthetic/[email protected][email protected] --output_dir ../../bucket/model_outputs/fr/syntax_modif_en/mlm --run_name fr_syntax_modif_en_500K_mlm &
  1. For composed transformations, apply multiple transformations by using multiple flags, e.g., --one_to_one_mapping --invert_word_order.
nohup python transformers/examples/xla_spawn.py --num_cores 8 transformers/examples/language-modeling/run_mlm_synthetic.py --warmup_steps 10000 --learning_rate 1e-4 --save_steps -1 --max_seq_length 512 --logging_steps 50 --overwrite_output_dir --model_type roberta --config_name config/en/roberta_8/config.json --tokenizer_name config/en/roberta_8/ --do_train --do_eval --max_steps 500000 --per_device_train_batch_size 16 --per_device_eval_batch_size 16 --train_file ../../bucket/pretrain_data/en/train.txt --validation_file ../../bucket/pretrain_data/en/valid.txt --output_dir ../../bucket/model_outputs/en/one_to_one_inverted/mlm --run_name en_one_to_one_inverted --one_to_one_mapping --invert_word_order --word_modification add &
  1. Using different domains for the original and derived language.
nohup python transformers/examples/xla_spawn.py --num_cores 8 transformers/examples/language-modeling/run_mlm_synthetic_transitive.py --warmup_steps 10000 --learning_rate 1e-4 --save_steps -1 --max_seq_length 512 --logging_steps 50 --overwrite_output_dir --model_type roberta --config_name config/en/roberta_8/config.json --tokenizer_name config/en/roberta_8/ --do_train --do_eval --max_steps 500000 --per_device_train_batch_size 16 --per_device_eval_batch_size 16 --train_file ../../bucket/pretrain_data/en/train_split_1.txt --transitive_file ../../bucket/pretrain_data/en/train_split_2.txt --validation_file ../../bucket/pretrain_data/en/valid.txt --output_dir ../../bucket/model_outputs/en/one_to_one_diff_source_100_more_steps/mlm --run_name en_one_to_one_diff_source_100_more_steps --one_to_one_mapping --word_modification add &

Fine-tuning and evaluation

This directory contains scripts used for downstream fine-tuning and evaluation.

  1. Transliteration, Inversion, and Permutation
  2. Syntax
  3. Composed transformations
  4. Using different domains for original and derived languages

Embedding alignment

Use this script to calculate embedding alignment for any model which uses Transliteration as one of the transformations.

Data and models

All the data used for our experiments, hosted on Google Cloud Bucket.

  1. Pre-training data - pretrain_data
  2. Downstream data - supervised_data
  3. Model files - model_outputs

Links to experiments and results

  1. Spreadsheets with run descriptions, commands, and weights and biases link
  2. Spreadsheet with all results
  3. Links to pre-training runs
  4. Link to fine-tuning and analysis

Citation

Please consider citing if you used our paper in your work!

To be updated soon!
Owner
Princeton Natural Language Processing
Princeton Natural Language Processing
GitHub Repository
Example for AUAV 2022 with obstacle avoidance.

AUAV 2022 Sample This is a sample PX4 based quadrotor path planning framework based on Ubuntu 20.04 and ROS noetic for the IEEE Autonomous UAS 2022 co

James Goppert 11 Sep 16, 2022
Implementation of a Transformer, but completely in Triton

Transformer in Triton (wip) Implementation of a Transformer, but completely in Triton. I'm completely new to lower-level neural net code, so this repo

Phil Wang 152 Dec 22, 2022
Pytorch implementation of SenFormer: Efficient Self-Ensemble Framework for Semantic Segmentation

SenFormer: Efficient Self-Ensemble Framework for Semantic Segmentation Efficient Self-Ensemble Framework for Semantic Segmentation by Walid Bousselham

61 Dec 26, 2022
U-Net implementation in PyTorch for FLAIR abnormality segmentation in brain MRI

U-Net for brain segmentation U-Net implementation in PyTorch for FLAIR abnormality segmentation in brain MRI based on a deep learning segmentation alg

562 Jan 02, 2023
Code for the paper: "On the Bottleneck of Graph Neural Networks and Its Practical Implications"

On the Bottleneck of Graph Neural Networks and its Practical Implications This is the official implementation of the paper: On the Bottleneck of Graph

75 Dec 22, 2022
An implementation of Equivariant e2 convolutional kernals into a convolutional self attention network, applied to radio astronomy data.

EquivariantSelfAttention An implementation of Equivariant e2 convolutional kernals into a convolutional self attention network, applied to radio astro

2 Nov 09, 2021
SpeechBrain is an open-source and all-in-one speech toolkit based on PyTorch.

The SpeechBrain Toolkit SpeechBrain is an open-source and all-in-one speech toolkit based on PyTorch. The goal is to create a single, flexible, and us

SpeechBrain 5.1k Jan 02, 2023
Official PyTorch implementation of "IntegralAction: Pose-driven Feature Integration for Robust Human Action Recognition in Videos", CVPRW 2021

IntegralAction: Pose-driven Feature Integration for Robust Human Action Recognition in Videos Introduction This repo is official PyTorch implementatio

Gyeongsik Moon 29 Sep 24, 2022
Code for the paper "VisualBERT: A Simple and Performant Baseline for Vision and Language"

This repository contains code for the following two papers: VisualBERT: A Simple and Performant Baseline for Vision and Language (arxiv) with a short

Natural Language Processing @UCLA 463 Dec 09, 2022
FTIR-Deep Learning - FTIR Deep Learning With Python

CANDIY-spectrum Human analyis of chemical spectra such as Mass Spectra (MS), Inf

Wei Mei 1 Jan 03, 2022
Code for the ICASSP-2021 paper: Continuous Speech Separation with Conformer.

Continuous Speech Separation with Conformer Introduction We examine the use of the Conformer architecture for continuous speech separation. Conformer

Sanyuan Chen (ι™ˆδΈ‰ε…ƒ) 81 Nov 28, 2022
SnapMix: Semantically Proportional Mixing for Augmenting Fine-grained Data (AAAI 2021)

SnapMix: Semantically Proportional Mixing for Augmenting Fine-grained Data (AAAI 2021) PyTorch implementation of SnapMix | paper Method Overview Cite

DavidHuang 126 Dec 30, 2022
This repo. is an implementation of ACFFNet, which is accepted for in Image and Vision Computing.

Attention-Guided-Contextual-Feature-Fusion-Network-for-Salient-Object-Detection This repo. is an implementation of ACFFNet, which is accepted for in I

5 Nov 21, 2022
Official implementation for paper: Feature-Style Encoder for Style-Based GAN Inversion

Feature-Style Encoder for Style-Based GAN Inversion Official implementation for paper: Feature-Style Encoder for Style-Based GAN Inversion. Code will

InterDigital 63 Jan 03, 2023
Sketch-Based 3D Exploration with Stacked Generative Adversarial Networks

pix2vox [Demonstration video] Sketch-Based 3D Exploration with Stacked Generative Adversarial Networks. Generated samples Single-category generation M

Takumi Moriya 232 Nov 14, 2022
a practicable framework used in Deep Learning. So far UDL only provide DCFNet implementation for the ICCV paper (Dynamic Cross Feature Fusion for Remote Sensing Pansharpening)

UDL UDL is a practicable framework used in Deep Learning (computer vision). Benchmark codes, results and models are available in UDL, please contact @

Xiao Wu 11 Sep 30, 2022
The code for our CVPR paper PISE: Person Image Synthesis and Editing with Decoupled GAN, Project Page, supp.

PISE The code for our CVPR paper PISE: Person Image Synthesis and Editing with Decoupled GAN, Project Page, supp. Requirement conda create -n pise pyt

jinszhang 110 Nov 21, 2022
Project of 'TBEFN: A Two-branch Exposure-fusion Network for Low-light Image Enhancement '

TBEFN: A Two-branch Exposure-fusion Network for Low-light Image Enhancement Codes for TMM20 paper "TBEFN: A Two-branch Exposure-fusion Network for Low

KUN LU 31 Nov 06, 2022
Indoor Panorama Planar 3D Reconstruction via Divide and Conquer

HV-plane reconstruction from a single 360 image Code for our paper in CVPR 2021: Indoor Panorama Planar 3D Reconstruction via Divide and Conquer (pape

sunset 36 Jan 03, 2023
Code and model benchmarks for "SEVIR : A Storm Event Imagery Dataset for Deep Learning Applications in Radar and Satellite Meteorology"

NeurIPS 2020 SEVIR Code for paper: SEVIR : A Storm Event Imagery Dataset for Deep Learning Applications in Radar and Satellite Meteorology Requirement

USAF - MIT Artificial Intelligence Accelerator 46 Dec 15, 2022