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
Torch implementation of SegNet and deconvolutional network

Torch implementation of SegNet and deconvolutional network

Fedor Chervinskii 5 Jul 17, 2020
TensorFlow (v2.7.0) benchmark results on an M1 Macbook Air 2020 laptop (macOS Monterey v12.1).

M1-tensorflow-benchmark TensorFlow (v2.7.0) benchmark results on an M1 Macbook Air 2020 laptop (macOS Monterey v12.1). I was initially testing if Tens

particle 2 Jan 05, 2022
Code for the Shortformer model, from the paper by Ofir Press, Noah A. Smith and Mike Lewis.

Shortformer This repository contains the code and the final checkpoint of the Shortformer model. This file explains how to run our experiments on the

Ofir Press 138 Apr 15, 2022
A Marvelous ChatBot implement using PyTorch.

PyTorch Marvelous ChatBot [Update] it's 2019 now, previously model can not catch up state-of-art now. So we just move towards the future a transformer

JinTian 223 Oct 18, 2022
Source code for Fixed-Point GAN for Cloud Detection

FCD: Fixed-Point GAN for Cloud Detection PyTorch source code of Nyborg & Assent (2020). Abstract The detection of clouds in satellite images is an ess

Joachim Nyborg 8 Dec 22, 2022
Official Pytorch Implementation of Unsupervised Image Denoising with Frequency Domain Knowledge

Unsupervised Image Denoising with Frequency Domain Knowledge (BMVC 2021 Oral) : Official Project Page This repository provides the official PyTorch im

Donggon Jang 12 Sep 26, 2022
Semantic code search implementation using Tensorflow framework and the source code data from the CodeSearchNet project

Semantic Code Search Semantic code search implementation using Tensorflow framework and the source code data from the CodeSearchNet project. The model

Chen Wu 24 Nov 29, 2022
Este conversor criará a medida exata para sua receita de capuccino gelado da grandiosa Rafaella Ballerini!

ConversorDeMedidas_CapuccinoGelado Este conversor criará a medida exata para sua receita de capuccino gelado da grandiosa Rafaella Ballerini! Requirem

Arthur Ottoni Ribeiro 48 Nov 15, 2022
PyImpetus is a Markov Blanket based feature subset selection algorithm that considers features both separately and together as a group in order to provide not just the best set of features but also the best combination of features

PyImpetus PyImpetus is a Markov Blanket based feature selection algorithm that selects a subset of features by considering their performance both indi

Atif Hassan 103 Dec 14, 2022
This repository contains the code used for the implementation of the paper "Probabilistic Regression with HuberDistributions"

Public_prob_regression_with_huber_distributions This repository contains the code used for the implementation of the paper "Probabilistic Regression w

David Mohlin 1 Dec 04, 2021
Raptor-Multi-Tool - Raptor Multi Tool With Python

Promises 🔥 20 Stars and I'll fix every error that there is 50 Stars and we will

Aran 44 Jan 04, 2023
Code for our NeurIPS 2021 paper Mining the Benefits of Two-stage and One-stage HOI Detection

CDN Code for our NeurIPS 2021 paper "Mining the Benefits of Two-stage and One-stage HOI Detection". Contributed by Aixi Zhang*, Yue Liao*, Si Liu, Mia

71 Dec 14, 2022
PyTorch implementation of the NIPS-17 paper "Poincaré Embeddings for Learning Hierarchical Representations"

Poincaré Embeddings for Learning Hierarchical Representations PyTorch implementation of Poincaré Embeddings for Learning Hierarchical Representations

Facebook Research 1.6k Dec 25, 2022
All course materials for the Zero to Mastery Machine Learning and Data Science course.

Zero to Mastery Machine Learning Welcome! This repository contains all of the code, notebooks, images and other materials related to the Zero to Maste

Daniel Bourke 1.6k Jan 08, 2023
Human Action Controller - A human action controller running on different platforms.

Human Action Controller (HAC) Goal A human action controller running on different platforms. Fun Easy-to-use Accurate Anywhere Fun Examples Mouse Cont

27 Jul 20, 2022
[CVPR2022] Representation Compensation Networks for Continual Semantic Segmentation

RCIL [CVPR2022] Representation Compensation Networks for Continual Semantic Segmentation Chang-Bin Zhang1, Jia-Wen Xiao1, Xialei Liu1, Ying-Cong Chen2

Chang-Bin Zhang 71 Dec 28, 2022
MVP Benchmark for Multi-View Partial Point Cloud Completion and Registration

MVP Benchmark: Multi-View Partial Point Clouds for Completion and Registration [NEWS] 2021-07-12 [NEW 🎉 ] The submission on Codalab starts! 2021-07-1

PL 93 Dec 21, 2022
For auto aligning, cropping, and scaling HR and LR images for training image based neural networks

ImgAlign For auto aligning, cropping, and scaling HR and LR images for training image based neural networks Usage Make sure OpenCV is installed, 'pip

15 Dec 04, 2022
Vision-Language Transformer and Query Generation for Referring Segmentation (ICCV 2021)

Vision-Language Transformer and Query Generation for Referring Segmentation Please consider citing our paper in your publications if the project helps

Henghui Ding 143 Dec 23, 2022
A framework to train language models to learn invariant representations.

Invariant Language Modeling Implementation of the training for invariant language models. Motivation Modern pretrained language models are critical co

6 Nov 16, 2022