Alexandre Bérard

Also published as: Alexandre Berard


2024

pdf bib
Understanding and Mitigating Language Confusion in LLMs
Kelly Marchisio | Wei-Yin Ko | Alexandre Berard | Théo Dehaze | Sebastian Ruder
Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing

We investigate a surprising limitation of LLMs: their inability to consistently generate text in a user’s desired language. We create the Language Confusion Benchmark (LCB) to evaluate such failures, covering 15 typologically diverse languages with existing and newly-created English and multilingual prompts. We evaluate a range of LLMs on monolingual and cross-lingual generation reflecting practical use cases, finding that Llama Instruct and Mistral models exhibit high degrees of language confusion and even the strongest models fail to consistently respond in the correct language. We observe that base and English-centric instruct models are more prone to language confusion, which is aggravated by complex prompts and high sampling temperatures. We find that language confusion can be partially mitigated via few-shot prompting, multilingual SFT and preference tuning. We release our language confusion benchmark, which serves as a first layer of efficient, scalable multilingual evaluation.

2023

pdf bib
Memory-efficient NLLB-200: Language-specific Expert Pruning of a Massively Multilingual Machine Translation Model
Yeskendir Koishekenov | Alexandre Berard | Vassilina Nikoulina
Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)

The recently released NLLB-200 is a set of multilingual Neural Machine Translation models that cover 202 languages. The largest model is based on a Mixture of Experts architecture and achieves SoTA results across many language pairs. It contains 54.5B parameters and requires at least four 32GB GPUs just for inference. In this work, we propose a pruning method that enables the removal of up to 80% of experts without further finetuning and with a negligible loss in translation quality, which makes it feasible to run the model on a single 32GB GPU. Further analysis suggests that our pruning metrics can identify language-specific experts.

pdf bib
NAVER LABS Europe’s Multilingual Speech Translation Systems for the IWSLT 2023 Low-Resource Track
Edward Gow-Smith | Alexandre Berard | Marcely Zanon Boito | Ioan Calapodescu
Proceedings of the 20th International Conference on Spoken Language Translation (IWSLT 2023)

This paper presents NAVER LABS Europe’s systems for Tamasheq-French and Quechua-Spanish speech translation in the IWSLT 2023 Low-Resource track. Our work attempts to maximize translation quality in low-resource settings using multilingual parameter-efficient solutions that leverage strong pre-trained models. Our primary submission for Tamasheq outperforms the previous state of the art by 7.5 BLEU points on the IWSLT 2022 test set, and achieves 23.6 BLEU on this year’s test set, outperforming the second best participant by 7.7 points. For Quechua, we also rank first and achieve 17.7 BLEU, despite having only two hours of translation data. Finally, we show that our proposed multilingual architecture is also competitive for high-resource languages, outperforming the best unconstrained submission to the IWSLT 2021 Multilingual track, despite using much less training data and compute.

2022

pdf bib
SMaLL-100: Introducing Shallow Multilingual Machine Translation Model for Low-Resource Languages
Alireza Mohammadshahi | Vassilina Nikoulina | Alexandre Berard | Caroline Brun | James Henderson | Laurent Besacier
Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing

In recent years, multilingual machine translation models have achieved promising performance on low-resource language pairs by sharing information between similar languages, thus enabling zero-shot translation. To overcome the “curse of multilinguality”, these models often opt for scaling up the number of parameters, which makes their use in resource-constrained environments challenging. We introduce SMaLL-100, a distilled version of the M2M-100(12B) model, a massively multilingual machine translation model covering 100 languages. We train SMaLL-100 with uniform sampling across all language pairs and therefore focus on preserving the performance of low-resource languages. We evaluate SMaLL-100 on different low-resource benchmarks: FLORES-101, Tatoeba, and TICO-19 and demonstrate that it outperforms previous massively multilingual models of comparable sizes (200-600M) while improving inference latency and memory usage. Additionally, our model achieves comparable results to M2M-100 (1.2B), while being 3.6x smaller and 4.3x faster at inference.

pdf bib
What Do Compressed Multilingual Machine Translation Models Forget?
Alireza Mohammadshahi | Vassilina Nikoulina | Alexandre Berard | Caroline Brun | James Henderson | Laurent Besacier
Findings of the Association for Computational Linguistics: EMNLP 2022

Recently, very large pre-trained models achieve state-of-the-art results in various natural language processing (NLP) tasks, but their size makes it more challenging to apply them in resource-constrained environments. Compression techniques allow to drastically reduce the size of the models and therefore their inference time with negligible impact on top-tier metrics. However, the general performance averaged across multiple tasks and/or languages may hide a drastic performance drop on under-represented features, which could result in the amplification of biases encoded by the models. In this work, we assess the impact of compression methods on Multilingual Neural Machine Translation models (MNMT) for various language groups, gender, and semantic biases by extensive analysis of compressed models on different machine translation benchmarks, i.e. FLORES-101, MT-Gender, and DiBiMT. We show that the performance of under-represented languages drops significantly, while the average BLEU metric only slightly decreases. Interestingly, the removal of noisy memorization with compression leads to a significant improvement for some medium-resource languages. Finally, we demonstrate that compression amplifies intrinsic gender and semantic biases, even in high-resource languages.

2021

pdf bib
Continual Learning in Multilingual NMT via Language-Specific Embeddings
Alexandre Berard
Proceedings of the Sixth Conference on Machine Translation

This paper proposes a technique for adding a new source or target language to an existing multilingual NMT model without re-training it on the initial set of languages. It consists in replacing the shared vocabulary with a small language-specific vocabulary and fine-tuning the new embeddings on the new language’s parallel data. Some additional language-specific components may be trained to improve performance (e.g., Transformer layers or adapter modules). Because the parameters of the original model are not modified, its performance on the initial languages does not degrade. We show on two sets of experiments (small-scale on TED Talks, and large-scale on ParaCrawl) that this approach performs as well or better as the more costly alternatives; and that it has excellent zero-shot performance: training on English-centric data is enough to translate between the new language and any of the initial languages.

pdf bib
Multilingual Domain Adaptation for NMT: Decoupling Language and Domain Information with Adapters
Asa Cooper Stickland | Alexandre Berard | Vassilina Nikoulina
Proceedings of the Sixth Conference on Machine Translation

Adapter layers are lightweight, learnable units inserted between transformer layers. Recent work explores using such layers for neural machine translation (NMT), to adapt pre-trained models to new domains or language pairs, training only a small set of parameters for each new setting (language pair or domain). In this work we study the compositionality of language and domain adapters in the context of Machine Translation. We aim to study, 1) parameter-efficient adaptation to multiple domains and languages simultaneously (full-resource scenario) and 2) cross-lingual transfer in domains where parallel data is unavailable for certain language pairs (partial-resource scenario). We find that in the partial resource scenario a naive combination of domain-specific and language-specific adapters often results in ‘catastrophic forgetting’ of the missing languages. We study other ways to combine the adapters to alleviate this issue and maximize cross-lingual transfer. With our best adapter combinations, we obtain improvements of 3-4 BLEU on average for source languages that do not have in-domain data. For target languages without in-domain data, we achieve a similar improvement by combining adapters with back-translation. Supplementary material is available at https://tinyurl.com/r66stbxj.

pdf bib
Multilingual Unsupervised Neural Machine Translation with Denoising Adapters
Ahmet Üstün | Alexandre Berard | Laurent Besacier | Matthias Gallé
Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing

We consider the problem of multilingual unsupervised machine translation, translating to and from languages that only have monolingual data by using auxiliary parallel language pairs. For this problem the standard procedure so far to leverage the monolingual data is _back-translation_, which is computationally costly and hard to tune. In this paper we propose instead to use _denoising adapters_, adapter layers with a denoising objective, on top of pre-trained mBART-50. In addition to the modularity and flexibility of such an approach we show that the resulting translations are on-par with back-translating as measured by BLEU, and furthermore it allows adding unseen languages incrementally.

pdf bib
Efficient Inference for Multilingual Neural Machine Translation
Alexandre Berard | Dain Lee | Stephane Clinchant | Kweonwoo Jung | Vassilina Nikoulina
Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing

Multilingual NMT has become an attractive solution for MT deployment in production. But to match bilingual quality, it comes at the cost of larger and slower models. In this work, we consider several ways to make multilingual NMT faster at inference without degrading its quality. We experiment with several “light decoder” architectures in two 20-language multi-parallel settings: small-scale on TED Talks and large-scale on ParaCrawl. Our experiments demonstrate that combining a shallow decoder with vocabulary filtering leads to almost 2 times faster inference with no loss in translation quality. We validate our findings with BLEU and chrF (on 380 language pairs), robustness evaluation and human evaluation.

2020

pdf bib
Naver Labs Europe’s Participation in the Robustness, Chat, and Biomedical Tasks at WMT 2020
Alexandre Berard | Ioan Calapodescu | Vassilina Nikoulina | Jerin Philip
Proceedings of the Fifth Conference on Machine Translation

This paper describes Naver Labs Europe’s participation in the Robustness, Chat, and Biomedical Translation tasks at WMT 2020. We propose a bidirectional German-English model that is multi-domain, robust to noise, and which can translate entire documents (or bilingual dialogues) at once. We use the same ensemble of such models as our primary submission to all three tasks and achieve competitive results. We also experiment with language model pre-training techniques and evaluate their impact on robustness to noise and out-of-domain translation. For German, Spanish, Italian, and French to English translation in the Biomedical Task, we also submit our recently released multilingual Covid19NMT model.

pdf bib
A Multilingual Neural Machine Translation Model for Biomedical Data
Alexandre Bérard | Zae Myung Kim | Vassilina Nikoulina | Eunjeong Lucy Park | Matthias Gallé
Proceedings of the 1st Workshop on NLP for COVID-19 (Part 2) at EMNLP 2020

We release a multilingual neural machine translation model, which can be used to translate text in the biomedical domain. The model can translate from 5 languages (French, German, Italian, Korean and Spanish) into English. It is trained with large amounts of generic and biomedical data, using domain tags. Our benchmarks show that it performs near state-of-the-art both on news (generic domain) and biomedical test sets, and that it outperforms the existing publicly released models. We believe that this release will help the large-scale multilingual analysis of the digital content of the COVID-19 crisis and of its effects on society, economy, and healthcare policies. We also release a test set of biomedical text for Korean-English. It consists of 758 sentences from official guidelines and recent papers, all about COVID-19.

pdf bib
Monolingual Adapters for Zero-Shot Neural Machine Translation
Jerin Philip | Alexandre Berard | Matthias Gallé | Laurent Besacier
Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)

We propose a novel adapter layer formalism for adapting multilingual models. They are more parameter-efficient than existing adapter layers while obtaining as good or better performance. The layers are specific to one language (as opposed to bilingual adapters) allowing to compose them and generalize to unseen language-pairs. In this zero-shot setting, they obtain a median improvement of +2.77 BLEU points over a strong 20-language multilingual Transformer baseline trained on TED talks.

2019

pdf bib
Machine Translation of Restaurant Reviews: New Corpus for Domain Adaptation and Robustness
Alexandre Berard | Ioan Calapodescu | Marc Dymetman | Claude Roux | Jean-Luc Meunier | Vassilina Nikoulina
Proceedings of the 3rd Workshop on Neural Generation and Translation

We share a French-English parallel corpus of Foursquare restaurant reviews, and define a new task to encourage research on Neural Machine Translation robustness and domain adaptation, in a real-world scenario where better-quality MT would be greatly beneficial. We discuss the challenges of such user-generated content, and train good baseline models that build upon the latest techniques for MT robustness. We also perform an extensive evaluation (automatic and human) that shows significant improvements over existing online systems. Finally, we propose task-specific metrics based on sentiment analysis or translation accuracy of domain-specific polysemous words.

pdf bib
Naver Labs Europe’s Systems for the Document-Level Generation and Translation Task at WNGT 2019
Fahimeh Saleh | Alexandre Berard | Ioan Calapodescu | Laurent Besacier
Proceedings of the 3rd Workshop on Neural Generation and Translation

Recently, neural models led to significant improvements in both machine translation (MT) and natural language generation tasks (NLG). However, generation of long descriptive summaries conditioned on structured data remains an open challenge. Likewise, MT that goes beyond sentence-level context is still an open issue (e.g., document-level MT or MT with metadata). To address these challenges, we propose to leverage data from both tasks and do transfer learning between MT, NLG, and MT with source-side metadata (MT+NLG). First, we train document-based MT systems with large amounts of parallel data. Then, we adapt these models to pure NLG and MT+NLG tasks by fine-tuning with smaller amounts of domain-specific data. This end-to-end NLG approach, without data selection and planning, outperforms the previous state of the art on the Rotowire NLG task. We participated to the “Document Generation and Translation” task at WNGT 2019, and ranked first in all tracks.

pdf bib
Naver Labs Europe’s Systems for the WMT19 Machine Translation Robustness Task
Alexandre Berard | Ioan Calapodescu | Claude Roux
Proceedings of the Fourth Conference on Machine Translation (Volume 2: Shared Task Papers, Day 1)

This paper describes the systems that we submitted to the WMT19 Machine Translation robustness task. This task aims to improve MT’s robustness to noise found on social media, like informal language, spelling mistakes and other orthographic variations. The organizers provide parallel data extracted from a social media website in two language pairs: French-English and Japanese-English (one for each language direction). The goal is to obtain the best scores on unseen test sets from the same source, according to automatic metrics (BLEU) and human evaluation. We propose one single and one ensemble system for each translation direction. Our ensemble models ranked first in all language pairs, according to BLEU evaluation. We discuss the pre-processing choices that we made, and present our solutions for robustness to noise and domain adaptation.

2017

pdf bib
LIG-CRIStAL Submission for the WMT 2017 Automatic Post-Editing Task
Alexandre Bérard | Laurent Besacier | Olivier Pietquin
Proceedings of the Second Conference on Machine Translation

2016

pdf bib
MultiVec: a Multilingual and Multilevel Representation Learning Toolkit for NLP
Alexandre Bérard | Christophe Servan | Olivier Pietquin | Laurent Besacier
Proceedings of the Tenth International Conference on Language Resources and Evaluation (LREC'16)

We present MultiVec, a new toolkit for computing continuous representations for text at different granularity levels (word-level or sequences of words). MultiVec includes word2vec’s features, paragraph vector (batch and online) and bivec for bilingual distributed representations. MultiVec also includes different distance measures between words and sequences of words. The toolkit is written in C++ and is aimed at being fast (in the same order of magnitude as word2vec), easy to use, and easy to extend. It has been evaluated on several NLP tasks: the analogical reasoning task, sentiment analysis, and crosslingual document classification.

pdf bib
Word2Vec vs DBnary: Augmenting METEOR using Vector Representations or Lexical Resources?
Christophe Servan | Alexandre Bérard | Zied Elloumi | Hervé Blanchon | Laurent Besacier
Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics: Technical Papers

This paper presents an approach combining lexico-semantic resources and distributed representations of words applied to the evaluation in machine translation (MT). This study is made through the enrichment of a well-known MT evaluation metric: METEOR. METEOR enables an approximate match (synonymy or morphological similarity) between an automatic and a reference translation. Our experiments are made in the framework of the Metrics task of WMT 2014. We show that distributed representations are a good alternative to lexico-semanticresources for MT evaluation and they can even bring interesting additional information. The augmented versions of METEOR, using vector representations, are made available on our Github page.