Challenge: Massively multilingual language models have shown strong performance in zero-shot (ZS-XLT) and few-shot cross-lingual transfer setups where models are fine-tuned on task data in a source language are transferred without any or with only a few annotated instances to the target language(s).
Approach: They propose a method that averages different checkpoints during task fine-tuning to improve model robustness.
Outcome: The proposed method overestimates model performance in cross-lingual transfer setups where models are evaluated at checkpoints that generalize best to validation instances in the target languages.

Similar Papers

Don’t Stop Fine-Tuning: On Training Regimes for Few-Shot Cross-Lingual Transfer with Multilingual Language Models (2022.emnlp-main)

Copied to clipboard

Challenge: Recent work highlights the fallacies of zero-shot cross-lingual transfer with large multilingual models.
Approach: They propose to replace sequential fine-tuning with joint fine-uning on source and target language instances.
Outcome: The proposed techniques yield improved and more stable FS-XLT across the board.
One For All & All For One: Bypassing Hyperparameter Tuning with Model Averaging for Cross-Lingual Transfer (2023.findings-emnlp)

Copied to clipboard

Challenge: Existing methods for zero-shot cross-lingual transfer are unreliable due to the lack of pretraining data.
Approach: They propose to accumulatively average model snapshots from different runs into a single model.
Outcome: The proposed protocol decouples performance maximization from hyperparameter tuning.
Why Does Zero-Shot Cross-Lingual Generation Fail? An Explanation and a Solution (2023.findings-acl)

Copied to clipboard

Challenge: Existing studies on cross-lingual transferability of multilingual LMs show that they can perform tasks in low-resource languages.
Approach: They propose a method to regularize the model from learning language invariant representations and a way to select model checkpoints without a development set in the target language.
Outcome: The proposed method reduces the accidental translation problem by 68% and improves the ROUGE-L score by 1.5 on average.
Key ingredients for effective zero-shot cross-lingual knowledge transfer in generative tasks (2024.naacl-long)

Copied to clipboard

Challenge: Existing studies have focused on zero-shot cross-lingual transfer . mBERT, mBART and mT5 provide high-quality representations for texts in various languages .
Approach: They propose to use mBART and NLLB-200 to finetune a multilingual pretrained language model on input-output pairs in one language and use it to make task predictions for inputs in other languages.
Outcome: The proposed approach significantly reduces generation in the wrong language with full finetuning and can be competitive in some cases.
From Zero to Hero: On the Limitations of Zero-Shot Language Transfer with Multilingual Transformers (2020.emnlp-main)

Copied to clipboard

Challenge: Existing studies show that multilingual transformers are less effective in resource-lean scenarios and for distant languages.
Approach: They propose to use massively multilingual transformers to pretrain languages . they show that MMTs are less effective in resource-lean scenarios and distant languages if they are pre-trained via language modeling .
Outcome: The proposed model is less effective in resource-lean scenarios and for distant languages than cross-lingual word embeddings.
Improving Zero-Shot Cross-Lingual Transfer Learning via Robust Training (2021.emnlp-main)

Copied to clipboard

Challenge: Pre-trained multilingual language encoders do not precisely align words and phrases across languages.
Approach: They propose a learning strategy for training robust models by drawing connections between adversarial examples and failure cases of zero-shot cross-lingual transfer.
Outcome: The proposed model can achieve good performance even if representations of different languages are not aligned well.
A Simple and Effective Method to Improve Zero-Shot Cross-Lingual Transfer Learning (2022.coling-1)

Copied to clipboard

Challenge: Existing zero-shot cross-lingual transfer methods rely on parallel corpora or bilingual dictionaries . however, its effect is limited by the gap between embedding clusters of different languages .
Approach: They propose Embedding-Push, Attention-Pull, and Robust targets to transfer English embeddings to virtual multilingual embedders without semantic loss.
Outcome: Experimental results show that the proposed method outperforms existing methods on cross-lingual tasks and can achieve a better multilingual alignment.
Analyzing the Evaluation of Cross-Lingual Knowledge Transfer in Multilingual Language Models (2024.eacl-long)

Copied to clipboard

Challenge: Recent advances in training multilingual models on large datasets have shown promising results in knowledge transfer across languages.
Approach: They challenge the assumption that high zero-shot performance reflects high cross-lingual ability by introducing more challenging setups involving instances with multiple languages.
Outcome: The proposed model can achieve high performance on multilingual benchmarks and on low-resource languages.
Massively Multilingual Transfer for NER (P19-1)

Copied to clipboard

Challenge: Existing approaches for cross-lingual transfer use a single source language, but there are exceptions.
Approach: They propose two techniques for modulating the transfer, suitable for zero-shot or few-shot learning, respectively.
Outcome: The proposed methods are much more effective than baseline models and rival oracle selection of the single best individual model.
Improving Massively Multilingual Neural Machine Translation and Zero-Shot Translation (2020.acl-main)

Copied to clipboard

Challenge: Existing approaches to improve multilingual neural machine translation (NMT) are weak, and lack robustness to support language pairs with varying typological characteristics.
Approach: They propose to deepen NMT models to support language pairs with varying typological characteristics by random online backtranslation.
Outcome: The proposed approach narrows the performance gap with bilingual models and improves zero-shot performance by 10 BLEU, approaching conventional pivot-based methods.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations