Pre-trained language models have been shown to substantially improve performance in many natural language tasks. Although the early focus of such models was single language pre-training, recent advances have resulted in cross-lingual and visual pre-training methods. In this paper, we combine these two approaches to learn visually-grounded cross-lingual representations. Specifically, we extend the translation language modelling (Lample and Conneau, 2019) with masked region classification, and perform pre-training with three-way parallel vision & language corpora. We show that when fine-tuned for multimodal machine translation, these models obtain state-of-the-art performance. We also provide qualitative insights into the usefulness of the learned grounded representations.
@inproceedings{caglayan-etal-2021-cross-lingual,
title = {{Cross-lingual Visual Pre-training for Multimodal Machine Translation}},
author = "Caglayan, Ozan and
Kuyu, Menekse and
Amac, Mustafa Sercan and
Madhyastha, Pranava and
Erdem, Erkut and
Erdem, Aykut and
Specia, Lucia",
booktitle = "Proceedings of the 16th Conference of the {E}uropean Chapter of the Association for Computational Linguistics: Short Papers",
month = apr,
year = "2021",
address = "online",
publisher = "Association for Computational Linguistics",
}
- The codebase is a revised, improved and extended version of XLM.
- No changes applied to multi-GPU code, which did not work well for us. All models were trained on a single GPU.
- Code portions that are not used are stripped to make the code simpler and clearer.
- Code paths for some of the original XLM tasks such as causal LM, back-translation, auto-encoding, etc. are not used/tested and probably broken due to refactoring.
If you need to experiment with pre-trained models or the fine-tuned NMTs/MMTs, you can download the set of checkpoints from here. Once extracted, the contents are as follows:
TLM: Pre-train on CC & fine-tune on Multi30k
- NMT:
finetuned-nmt-over-tlm - MMT:
finetuned-mmt-over-tlm
VTLM: Pre-train on CC & fine-tune on Multi30k
- NMT:
finetuned-nmt-over-vtlm - MMT:
finetuned-mmt-over-vtlm
VTLM: Alternative (0% visual masking during pre-training)
- MMT:
finetuned-mmt-over-vtlm-alternative
(Each folder above contains 3 runs of pre-training and hence, 3 checkpoint files.)
The three pre-trained checkpoints used for the above models are included
in the folder pretrained-checkpoints.
We made some of the datasets and features accessible through zenodo.org.
Please run scripts/00-download.sh from within the data/ subfolder to download:
- Conceptual Captions (CC) dataset with German automatic translations
- Tensorflow Object detection checkpoint folder
- EN-DE corpus for Multi30k dataset and pre-extracted object features
$ cd data
$ scripts/00-download.sh
- You will need to fetch all the images provided as URLs in the
conceptual_captions/cc-en-de.tsv.*files. However, it is very likely that some of the images are no longer accessible, breaking the train, valid, test splits provided by this paper. We still suggest to keep those splits as it is, especially thetestset, if you would like to replicate the results in this paper. - We are planning to at least provide the pre-extracted features for the test set.
- A sample script to download the images is provided in
scripts/02-download-images.sh. - You need TensorFlow >= 2 to extract the features from the downloaded images. The script
for this is
scripts/99-feature-extractor.py. You can divide the image lists into as many sub-files as there are free GPUs on your machine, to parallelize the feature extraction. NOTE: You need to pass--poolargument to this script as the codebase expects a feature vector per each image rather than a 3D convolutional map.
After the downloading step, run scripts/01-process-cc.sh from within the data/ subfolder
to tokenize and BPE-ize the Conceptual Captions dataset files. The dataset is already
splitted into train, dev and test sets. The produced .pth files will be used by the
code to load the binarized dataset splits into memory efficiently.
Manually rework the files in order to filter out samples for which the images are no longer accessible, and re-run the script above until everything is smooth.
After the downloading step, run scripts/01-process-multi30k.sh from within the data/ subfolder
to tokenize and BPE-ize the Multi30k dataset files. The test set in our case here
will refer to test_2016_flickr split.
At inference time, you need to manually switch the test.de-en.*pth and image order files
to decode other test sets. A hacky think to do is to run the processing script multiple
times to create copies of folders where each folder has a different test set.
We provide shell scripts under scripts/ for pre-training and fine-tuning workflows that are
used throughout the paper. For pre-training the TLM and VTLM, we provide two
toy examples that'll do pre-training on Multi30k as a starting point:
train-tlm-on-multi30k.sh: Trains a TLM model on the Multi30k En-De corpus.train-vtlm-on-multi30k.sh: Trains a VTLM model on the Multi30k En-De corpus.
For pre-training TLM and VTLM on Conceptual Captions (CC), you need
to download CC images and extract object detection features. Once
this is done, you could use train-tlm-on-cc.sh and train-vtlm-on-cc.sh scripts.
You can use train-[nmt|mmt]-from-scratch-on-multi30k.sh scripts to train NMT/MMT
baselines without pre-training / fine-tuning.
You can use finetune-[nmt|mmt]-multi30k-over-[tlm|vtlm].sh scripts to fine-tune
existing TLM/VTLM checkpoints on NMT and MMT tasks, using the En-De Multi30k corpus.
decode-[nmt|mmt]-multi30k.sh scripts can be used on arbitrarily trained NMT/MMT
checkpoints, to decode translations of val and test_2016_flickr test set.
By default, it uses a beam size of 8.
Several probes-*.sh scripts are provided to reproduce the masking experiments
of the paper.