Jump to Content
Orhan Firat

Orhan Firat

Authored Publications
Google Publications
Other Publications
Sort By
  • Title
  • Title, desc
  • Year
  • Year, desc
    Preview abstract Automatic evaluation of machine translation (MT) is a critical tool driving the rapid iterative development of MT systems. While considerable progress has been made on direct estimation of quality scores, the resulting metrics lack the informativeness of more detailed schemes that annotate individual errors, such as Multidimensional Quality Metrics (MQM). In this paper, we fill this gap by proposing \textbf{\textsc{AutoMQM}}, a prompting technique which leverages the \textit{reasoning} and \textit{in-context learning} capabilities of large language models (LLMs) and asks them to identify and categorize errors in translations. We start by evaluating recent LLMs, such as PaLM and PaLM-2, through simple \textit{score prediction} prompting, and we study the impact of labeled data through in-context learning and finetuning. We then evaluate \textsc{AutoMQM} with PaLM-2 models, and we find that it improves performance compared to just prompting for scores (with particularly large gains for larger models) while providing interpretability through error spans that align with human annotations. View details
    Preview abstract We present FRMT, a new dataset and evaluation benchmark for Few-shot Region-aware Machine Translation, a type of style-targeted translation. The dataset consists of professional translations from English into two regional variants each of Portuguese and Mandarin Chinese. Source documents are selected to enable detailed analysis of phenomena of interest, including lexically distinct terms and distractor terms. We explore automatic evaluation metrics for FRMT and validate their correlation with expert human evaluation across both region-matched and mismatched rating scenarios. Finally, we present a number of baseline models for this task, and offer guidelines for how researchers can train, evaluate, and compare their own models. Our dataset and evaluation code are publicly available: https://bit.ly/frmt-task View details
    Preview abstract Neural machine translation (NMT) has progressed rapidly over the past several years, and modern models are able to achieve relatively high quality using only monolingual text data, an approach dubbed Unsupervised Machine Translation, or UNMT. However, these models still struggle in a variety of ways, including aspects of translation that for a human are the easiest---for instance, correctly translating common nouns. This work explores a cheap and abundant resource to combat this problem: bilingual lexicons (\textsc{BiLex}s). We test the efficacy of bilingual lexicons in a real-world set-up, on 200-language translation models trained on web-mined text. We present several findings: (1) we demonstrate the most effective ways to use this resource for MT by extensively experimenting with lexical data augmentation techniques, such as codeswitching and lexical prompting; (2) we pinpoint what settings and languages are benefited most from lexical data augmentation; and (3) we provide an empirical, per-language analysis of the quality of the public resource PanLex, a multilingual lexicon covering thousands of languages. View details
    Preview abstract In this work, we study the evolution of the loss Hessian across many classification tasks in order to understand the effect the curvature of the loss has on the training dynamics. Whereas prior work has focused on how different learning rates affect the loss Hessian observed during training, we also analyze the effects of model initialization, architectural choices, and common training heuristics such as gradient clipping and learning rate warmup. Our results demonstrate that successful model and hyperparameter choices allow the early optimization trajectory to either avoid---or navigate out of---regions of high curvature and into flatter regions that tolerate a higher learning rate. Our results suggest a unifying perspective on how disparate mitigation strategies for training instability ultimately address the same underlying failure mode of neural network optimization, namely poor conditioning. Inspired by the conditioning perspective, we show that learning rate warmup can improve training stability just as much as batch normalization, layer normalization, MetaInit, GradInit, and Fixup initialization. View details
    Preview abstract We introduce \xtremes, a new benchmark to evaluate universal cross-lingual speech representations in many languages. XTREME-S covers four task families: speech recognition, classification, retrieval and speech-to-text translation. Covering 102 languages from 10+ language families, 3 different domains and 4 task families, XTREME-S aims to simplify multilingual speech representation evaluation, as well as catalyze research in ``universal'' speech representation learning. This paper describes the new benchmark and establishes the first speech-only and speech-text baselines using XLS-R and mSLAM on all downstream tasks. We motivate the design choices and detail how to use the benchmark. The code and pre-processing scripts will be made publicly available.\footnote{\small\url{https://huggingface.co/datasets/google/xtreme_s}} View details
    PaLM: Scaling Language Modeling with Pathways
    Sharan Narang
    Jacob Devlin
    Maarten Bosma
    Hyung Won Chung
    Sebastian Gehrmann
    Parker Schuh
    Sasha Tsvyashchenko
    Abhishek Rao
    Yi Tay
    Noam Shazeer
    Nan Du
    Reiner Pope
    James Bradbury
    Guy Gur-Ari
    Toju Duke
    Henryk Michalewski
    Xavier Garcia
    Liam Fedus
    David Luan
    Barret Zoph
    Ryan Sepassi
    David Dohan
    Shivani Agrawal
    Mark Omernick
    Marie Pellat
    Aitor Lewkowycz
    Erica Moreira
    Rewon Child
    Oleksandr Polozov
    Zongwei Zhou
    Michele Catasta
    Jason Wei
    arxiv:2204.02311 (2022)
    Preview abstract Large language models have been shown to achieve remarkable performance across a variety of natural language tasks using few-shot learning, which drastically reduces the number of task-specific training examples needed to adapt the model to a particular application. To further our understanding of the impact of scale on few-shot learning, we trained a 540-billion parameter, densely activated, Transformer language model, which we call Pathways Language Model PaLM. We trained PaLM on 6144 TPU v4 chips using Pathways, a new ML system which enables highly efficient training across multiple TPU Pods. We demonstrate continued benefits of scaling by achieving state-of-the-art few-shot learning results on hundreds of language understanding and generation benchmarks. On a number of these tasks, PaLM 540B achieves breakthrough performance, outperforming the finetuned state-of-the-art on a suite of multi-step reasoning tasks, and outperforming average human performance on the recently released BIG-bench benchmark. A significant number of BIG-bench tasks showed discontinuous improvements from model scale, meaning that performance steeply increased as we scaled to our largest model. PaLM also has strong capabilities in multilingual tasks and source code generation, which we demonstrate on a wide array of benchmarks. We additionally provide a comprehensive analysis on bias and toxicity, and study the extent of training data memorization with respect to model scale. Finally, we discuss the ethical considerations related to large language models and discuss potential mitigation strategies. View details
    Preview abstract In this paper we share findings from our effort towards building practical machine translation (MT) systems capable of translating across over one thousand languages. We describe results across three research domains: (i) Building clean, web-mined datasets by leveraging semi-supervised pre-training for language-id and developing data-driven filtering techniques; (ii) Leveraging massively multilingual MT models trained with supervised parallel data for over $100$ languages and small monolingual datasets for over $1000$ languages to enable translation for several previously under-studied languages; and (iii) Studying the limitations of evaluation metrics for long tail languages and conducting qualitative analysis of the outputs from our MT models. We hope that our work provides useful insights to practitioners working towards building MT systems for long tail languages, and highlights research directions that can complement the weaknesses of massively multilingual pre-trained models in data-sparse settings. View details
    Sparsely Activated Language Models are Efficient In-Context Learners
    Barret Richard Zoph
    Dmitry (Dima) Lepikhin
    Emma Wang
    Kun Zhang
    Liam B. Fedus
    Maarten Paul Bosma
    Marie Pellat
    Maxim Krikun
    Nan Du
    Simon Tong
    Tao Wang
    Toju Duke
    Yuanzhong Xu
    Zongwei Zhou
    (2022)
    Preview abstract Scaling language models with more data, compute and parameters has driven significant progress in natural language processing. For example, thanks to scaling, GPT-3 was able to achieve strong performance on few-shot learning. However, training these large dense models require significant amounts of computing resources. In this paper, we develop a family of sparsely activated mixture-of-expert language models named \glam (\textbf{G}eneralist \textbf{La}nguage \textbf{M}odel), which can have many more parameters but require significant less training cost than dense models. The largest \glam has 1.2 trillion parameters, which is approximately 7x larger than GPT-3 but can be trained more efficiently. With only 1/3 of energy consumption to train GPT-3, \glam achieves better overall performance on 29 zero-shot and one-shot NLP tasks. For example, \glam gets 75.0\% one-shot exact match accuracy on the TriviaQA test server, a significant improvement over 68.0\% obtained by GPT-3. View details
    Preview abstract Multilingual neural machine translation (NMT) typically learns to maximize the likelihood of training examples from a combination set of multiple language pairs. However, this mechanical combination only relies on the basic sharing to learn the inductive bias, which undermines the generalization and transferability of multilingual NMT models. In this paper, we introduce a multilingual crossover encoder-decoder (mXEnDec) to fuse language pairs at instance level to exploit cross-lingual signals. For better fusions on multilingual data, we propose several techniques to deal with the language interpolation, dissimilar language fusion and heavy data imbalance. Experimental results on a large-scale WMT multilingual data set show that our approach significantly improves model performance on general multilingual test sets and the model transferability on zero-shot test sets (up to $+5.53$ BLEU). Results on noisy inputs demonstrates the capability of our approach to improve model robustness against the code-switching noise. We also conduct qualitative and quantitative representation comparisons to analyze the advantages of our approach at the representation level. View details
    Quality at a Glance: An Audit of Web-Crawled Multilingual Datasets
    Julia Kreutzer
    Lisa Wang
    Ahsan Wahab
    Nasanbayar Ulzii-Orshikh
    Allahsera Auguste Tapo
    Nishant Subramani
    Artem Sokolov
    Claytone Sikasote
    Monang Setyawan
    Supheakmungkol Sarin
    Sokhar Samb
    Benoît Sagot
    Clara E. Rivera
    Annette Rios
    Isabel Papadimitriou
    Salomey Osei
    Pedro Javier Ortiz Suárez
    Iroro Fred Ọ̀nọ̀mẹ̀ Orife
    Kelechi Ogueji
    Rubungo Andre Niyongabo
    Toan Nguyen
    Mathias Müller
    André Müller
    Shamsuddeen Hassan Muhammad
    Nanda Muhammad
    Ayanda Mnyakeni
    Jamshidbek Mirzakhalov
    Tapiwanashe Matangira
    Colin Leong
    Nze Lawson
    Yacine Jernite
    Mathias Jenny
    Bonaventure F. P. Dossou
    Sakhile Dlamini
    Nisansa de Silva
    Sakine Çabuk Ballı
    Stella Biderman
    Alessia Battisti
    Ahmed Baruwa
    Pallavi Baljekar
    Israel Abebe Azime
    Ayodele Awokoya
    Duygu Ataman
    Orevaoghene Ahia
    Oghenefego Ahia
    Sweta Agrawal
    Mofetoluwa Adeyemi
    TACL (2022)
    Preview abstract With the success of large-scale pre-training and multilingual modeling in Natural Language Processing (NLP), recent years have seen a proliferation of large, web-mined text datasets covering hundreds of languages. However, to date there has been no systematic analysis of the quality of these publicly available datasets, or whether the datasets actually contain content in the languages they claim to represent. In this work, we manually audit the quality of 205 language-specific corpora released with five major public datasets (CCAligned, ParaCrawl, WikiMatrix, OSCAR, mC4), and audit the correctness of language codes in a sixth (JW300). We find that lower-resource corpora have systematic issues: at least 15 corpora are completely erroneous, and a significant fraction contains less than 50% sentences of acceptable quality. Similarly, we find 82 corpora that are mislabeled or use nonstandard/ambiguous language codes. We demonstrate that these issues are easy to detect even for non-speakers of the languages in question, and supplement the human judgements with automatic analyses. Inspired by our analysis, we recommend techniques to evaluate and improve multilingual corpora and discuss the risks that come with low-quality data releases. View details
    Image Translation Network
    Puneet Jain
    Qi Ge
    Sihang Liang
    Image Translation Model (2021)
    Preview abstract We present an end-to-end neural network to translate images containing text from one language to another. Traditionally, a cascaded approach of optical character recognition (OCR) followed by neural machine translation (NMT) is used to solve this problem. However, the cascaded approach compounds OCR and NMT errors, and incurs longer latency, performs poorly in multiline cases. Our simplified approach combines OCR and NMT into one end-to-end model. Our neural architecture follows the encoder-decoder paradigm, with a convolutional encoder and an autoregressive Transformer decoder. Trained end-to-end, our proposed model yields significant improvements on multiple dimensions, (i) achieves higher translation accuracy due to better error propagation, (ii) incurs lower inference latency due to smaller network size, and (iii) translates multiline paragraphs and understands reading order of the lines, (iv) eliminates source side vocabulary. We train several variations of encoders and decoders on a synthetic corpus of 120M+ English-French images and show that our approach outperforms the cascaded approach with a large margin in both the automatic metrics and the detailed side-by-side human evaluation. View details
    Beyond Distillation: Task-level Mixture-of-Experts for Efficient Inference
    Dmitry (Dima) Lepikhin
    Maxim Krikun
    Beyond Distillation: Task-level Mixture-of-Experts for Efficient Inference (2021)
    Preview abstract Sparse Mixture-of-Experts (MoE) has been a successful approach for scaling multilingual translation models to billions of parameters without a proportional increase in training computation. However, MoE models are prohibitively large and practitioners often resort to methods such as distillation for serving. In this work, we investigate routing strategies at different granularity (token, sentence, task) in MoE models to bypass distillation. Experiments on WMT and a web-scale dataset suggest that task-level routing (task-MoE) enables us to extract smaller, ready-to-deploy sub-networks from large sparse models. On WMT, our task-MoE with 32 experts (533M parameters) outperforms the best performing token-level MoE model (token-MoE) by +1.0 BLEU on average across 30 language pairs. The peak inference throughput is also improved by a factor of 1.9x when we route by tasks instead of tokens. While distilling a token-MoE to a smaller dense model preserves only 32% of the BLEU gains, our sub-network task-MoE, by design, preserves all the gains with the same inference cost as the distilled student model. Finally, when scaling up to 200 language pairs, our 128-expert task-MoE (13B parameters) performs competitively with a token-level counterpart, while improving the peak inference throughput by a factor of 2.6x. View details
    Preview abstract Massively multilingual models subsuming tens or even hundreds of languages pose great challenges to multi-task optimization. While it is a common practice to apply a language-agnostic procedure optimizing a joint multilingual task objective, how to properly characterize and take advantage of its underlying problem structure for improving optimization efficiency remains under-explored. In this paper, we attempt to peek into the black-box of multilingual optimization through the lens of loss function geometry. We find that gradient similarity measured along the optimization trajectory is an important signal, which correlates well with not only language proximity but also the overall model performance. Such observation helps us to identify a critical limitation of existing gradient-based multi-task learning methods, and thus we derive a simple and scalable optimization procedure, named Gradient Vaccine, which encourages more geometrically aligned parameter updates for close tasks. Empirically, our method obtains significant model performance gains on multilingual machine translation and XTREME benchmark tasks for multilingual language models. Our work reveals the importance of properly measuring and utilizing language proximity in multilingual optimization, and has broader implications for multi-task learning beyond multilingual modeling. View details
    Preview abstract Over the last few years two promising research directions in low-resource neural machine translation (NMT) have emerged. The first focuses on utilizing high-resource languages to improve the quality of low-resource languages via multilingual NMT. The second direction employs monolingual data with self-supervision to pre-train translation models, followed by fine-tuning on small amounts of supervised data. In this work, we join these two lines of research and demonstrate the efficacy of monolingual data with self-supervision in multilingual NMT. We offer three major results: (i) Using monolingual data significantly boosts the translation quality of low-resource languages in multilingual models. (ii) Self-supervision improves zero-shot translation quality in multilingual models. (iii) Leveraging monolingual data with self-supervision provides a viable path towards adding new languages to multilingual models, getting up to 28 BLEU on ro-en translation without any parallel data or back-translation. View details
    Complete Multilingual Neural Machine Translation
    Proceedings of the Fifth Conference on Machine Translation (Volume 1: Research Papers) (2020)
    Preview abstract Multilingual Neural Machine Translation (MNMT) models are commonly trained on a joint set of bilingual corpora which is acutely English-centric (i.e. English either as the source or target language). While direct data between two languages that are non-English is explicitly available at times, its use is not common. In this paper, we first take a step back and look at the commonly used bilingual corpora (WMT), and resurface the existence and importance of implicit structure that existed in it: multi-way alignment across examples (the same sentence in more than two languages). We set out to study the use of multi-way aligned examples to enrich the original English-centric parallel corpora. We reintroduce this direct parallel data from multi-way aligned corpora between all source and target languages. By doing so, the English-centric graph expands into a complete graph, every language pair being connected. We call MNMT with such connectivity pattern complete Multilingual Neural Machine Translation (cMNMT) and demonstrate its utility and efficacy with a series of experiments and analysis. In combination with a novel training data sampling strategy that is conditioned on the target language only, cMNMT yields competitive translation quality for all language pairs. We further study the size effect of multi-way aligned data, its transfer learning capabilities and how it eases adding a new language in MNMT. Finally, we stress test cMNMT at scale and demonstrate that we can train a cMNMT model with up to 111*112=12,432 language pairs that provides competitive translation quality for all language pairs. View details
    Preview abstract Recently proposed Massively Multilingual Neural Machine Translation system has been shown to be capable of translating 102 languages to and from English within a single model. In this paper, we evaluate the cross-lingual effectiveness of representations from the encoder of such a model on 5 downstream classification and sequence tagging tasks spanning more than 50 languages. We compare our results to a strong multilingual baseline, BERT and show modest gains on zero-shot cross-lingual transfer in 4 out of these 5 tasks. Our results provide strong insight into how applicable the representations learned from multilingual machine translation are, across languages and tasks. View details
    Towards End-to-End In-Image Neural Machine Translation
    Elman Mansimov
    Jakob Uszkoreit
    Mitchell Stern
    Puneet Jain
    EMNLP, NLP Beyond Text workshop, 2020 (2020)
    Preview abstract In this paper, we offer a preliminary investigation into the task of in-image machine translation: transforming an image containing text in one language into an image containing the same text in another language. We propose an end-to-end neural model for this task inspired by recent approaches to neural machine translation, and demonstrate promising initial results based purely on pixel-level supervision. We then offer a qualitative evaluation of our system outputs and discuss some common failure modes. Finally, we conclude with directions for future work. View details
    Preview abstract Much recent progress in applications of machine learning models to NLP has been driven by benchmarks that evaluate models across a wide variety of tasks. However, these broad-coverage benchmarks have been mostly limited to English, and despite an increasing interest in multilingual models, a benchmark that enables the comprehensive evaluation of such methods on a diverse range of languages and tasks is still missing. To this end, we introduce the Cross-lingual TRansfer Evaluation of Multilingual Encoders benchmark, a multi-task benchmark for evaluating the cross-lingual generalization capabilities of multilingual representations across 40 languages and 9 tasks. We demonstrate that while models tested on English reach human performance on many tasks, there is still a sizable gap in the performance of cross-lingually transferred models, particularly on syntactic and sentence retrieval tasks. There is also a wide spread of results across languages. We will release the benchmark to encourage research on cross-lingual learning methods that transfer linguistic knowledge across a diverse and representative set of languages and tasks. View details
    GShard: Scaling Giant Models With Conditional Computation and Automatic Sharding
    Dehao Chen
    Dmitry (Dima) Lepikhin
    HyoukJoong Lee
    Maxim Krikun
    Noam Shazeer
    Yuanzhong Xu
    ICLR 2021 (2020) (to appear)
    Preview abstract Neural network scaling has been critical for improving the model quality in many real world machine learning applications with vast amounts of training data and compute. Although this trend of scaling is affirmed to be a sure-fire approach for better model quality, there are challenges on the path such as the computation cost, ease of programming, and efficient implementation on parallel devices. GShard is a module composed of a set of lightweight annotation APIs and an extension to the XLA compiler. It provides an elegant way to express a wide range of parallel computation patterns with minimal changes of existing model code. It enabled us to scale up multilingual machine translation Transformer model with Sparsely-Gated Mixture-of-Experts beyond 600 billion parameters using automatic sharding. We demonstrate that such a giant model can be easily trained on 2048 TPU v3 accelerators in 4 days to achieve far superior quality for translation from 100 languages to English compared to the prior art. View details
    Preview abstract Neural Networks trained with gradient descent are known to be susceptible to catastrophic forgetting caused by parameter shift during the training process. In the context of Neural Machine Translation (NMT) this results in poor performance on heterogeneous datasets, and on sub-tasks like rare phrase translation. On the other hand, non-parametric approaches are immune to forgetting, perfectly complementing the generalization ability of NMT. However, attempts to combine non-parametric or retrieval based approaches with NMT have only been successful on narrow domains, possibly due to over-reliance on sentence level retrieval. We propose a novel n-gram level retrieval approach that relies on local phrase level similarities, allowing us to retrieve neighbors that are useful for translation even when overall sentence similarity is low. We complement this with an expressive neural model, allowing our model to extract information from the noisy retrieved context. We evaluate our semi-parametric NMT approach on a heterogeneous dataset composed of WMT, IWSLT, JRC-Acquis and OpenSubtitles, and demonstrate gains on all 4 evaluation sets. The semi-parametric nature of our approach also opens the door for non-parametric domain adaptation, demonstrating strong inference-time adaptation performance on new domains without the need for any parameter updates. View details
    Preview abstract We introduce our efforts towards building a universal neural machine translation (NMT) system capable of translating between any language pair. We set a milestone towards this goal by building a single massively multilingual NMT model handling 103 languages trained over 25 billion examples. Our system demonstrates effective transfer learning ability, significantly improving translation quality of low-resource languages, while keeping high-resource language translation quality on-par with competitive bilingual baselines. We provide in-depth analysis of various aspects of model building that are crucial to the quality and practicality towards universal NMT. While we prototype a high-quality universal translation system, our extensive empirical analysis exposes issues that need to be further addressed, and we suggest directions for future research. View details
    Preview abstract Multilingual Neural Machine Translation (NMT) models have yielded large empirical success in transfer learning settings. However, these black-box representations are poorly understood, and their mode of transfer remains elusive. In this work, we attempt to understand massively multilingual NMT representations (with over 100 languages) using Singular Value Canonical Correlation Analysis (SVCCA), a representation similarity framework that allows us to compare representations across different languages, layers and models. Our analysis validates several empirical results and long-standing intuitions, and unveils new observations regarding how representations evolve in a multilingual translation model. We draw two major results from our analysis: (i) Representations of the same sentences across different languages cluster based on linguistic similarity and (ii) Source sentence representations learned by the encoder are dependent on the target language. We further confirm our observations with carefully designed experiments and connect our findings with existing results in multilingual NMT and cross-lingual transfer learning. View details
    Massively Multilingual Neural Machine Translation
    Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), Association for Computational Linguistics, Minneapolis, Minnesota, pp. 3874-3884 (to appear)
    Preview abstract Multilingual Neural Machine Translation enables training a single model that supports translation from multiple source languages into multiple target languages. We perform extensive experiments in training massively multilingual NMT models, involving up to 103 distinct languages and 204 translation directions simultaneously. We explore different setups for training such models and analyze the trade-offs between translation quality and various modeling decisions. We report results on the publicly available TED talks multilingual corpus where we show that massively multilingual many-to-many models are effective in low resource settings, outperforming the previous state-of-the-art while supporting up to 59 languages in 116 translation directions in a single model. Our experiments on a large-scale dataset with 103 languages, 204 trained directions and one million examples per direction also show promising results, surpassing strong bilingual baselines and encouraging future work on massively multilingual NMT. View details
    Preview abstract Scaling up deep neural network capacity has been known as an effective approach to improving model quality for several different machine learning tasks. In many cases, increasing model capacity beyond the memory limit of a single accelerator has required developing special algorithms or infrastructure. These solutions are often architecture-specific and do not transfer to other tasks. To address the need for efficient and task-independent model parallelism, we introduce GPipe, a pipeline parallelism library that allows scaling any network that can be expressed as a sequence of layers. By pipelining different sub-sequences of layers on separate accelerators, GPipe provides the flexibility of scaling a variety of different networks to gigantic sizes efficiently. Moreover, GPipe utilizes a novel batch-splitting pipelining algorithm, resulting in almost linear speedup when a model is partitioned across multiple accelerators. We demonstrate the advantages of GPipe by training large-scale neural networks on two different tasks with distinct network architectures: (i) Image Classification: We train a 557-million-parameter AmoebaNet model and attain a top-1 accuracy of 84.4% on ImageNet-2012, (ii) Multilingual Neural Machine Translation: We train a single 6-billion-parameter, 128-layer Transformer model on a corpus spanning over 100 languages and achieve better quality than all bilingual models. View details
    Preview abstract Translating characters instead of words or word-fragments has the potential to simplify the processing pipeline for neural machine translation (NMT), and improve results by eliminating hyper-parameters and manual feature engineering. However, it results in longer sequences in which each symbol contains less information, creating both modeling and computational challenges. In this paper, we show that the modeling problem can be solved by standard sequence-to-sequence architectures of sufficient depth, and that deep models operating at the character level outperform identical models operating over word fragments. This result implies that alternative architectures for handling character input are better viewed as methods for reducing computation time than as improved ways of modeling longer sequences. From this perspective, we evaluate several techniques for character-level NMT, verify that they do not match the performance of our deep character baseline model, and evaluate the performance versus computation time tradeoffs they offer. Within this framework, we also perform the first evaluation for NMT of conditional computation over time, in which the model learns which timesteps can be skipped, rather than having them be dictated by a fixed schedule specified before training begins. View details
    Preview abstract The past year has witnessed rapid advances in sequence-to-sequence (seq2seq) modeling for Machine Translation (MT). The classic RNN-based approaches to MT were first out-performed by the convolutional seq2seq model, which was then out-performed by the more recent Transformer model. Each of these new approaches consists of a fundamental architecture accompanied by a set of modeling and training techniques that are in principle applicable to other seq2seq architectures. In this paper, we tease apart the new architectures and their accompanying techniques in two ways. First, we identify several key modeling and training techniques, and apply them to the RNN architecture, yielding a new RNMT+ model that outperforms all of the three fundamental architectures on the benchmark WMT'14 English to French and English to German tasks. Second, we analyze the properties of each fundamental seq2seq architecture and devise new hybrid architectures intended to combine their strengths. Our hybrid models obtain further improvements, outperforming the RNMT+ model on both benchmark datasets. View details
    Preview abstract While current state-of-the-art NMT models, both LSTM based and Transformers, are much deeper compared to their early counterparts, they are still shallow in comparison to convolutional models used for both text and vision applications. In this work we attempt to train significantly (2-3x) deeper transformer and BiLSTM encoders for machine translation. We propose a simple modification to the attention mechanism that eases the optimization of deeper models, and results in significant improvements on the benchmark WMT'14 English-German and WMT'15 Czech-English tasks for both architectures. View details
    Hallucinations in Neural Machine Translation
    Ashish Agarwal
    Clara Wong-Fannjiang
    David Sussillo
    ICLR (2018) (to appear)
    Preview abstract Neural machine translation (NMT) systems have reached state of the art performance in translating text and are in wide deployment. Yet little is understood about how these systems function or how they break. Here we show that NMT systems are susceptible to producing highly pathological translations that are completely untethered from the source material, which we term {\it hallucinations}. Such pathological translations are problematic because they are are deeply disturbing of user trust and are easy to find with a simple search. We describe a method to generate hallucinations and show that many common variations of the NMT architecture are susceptible to them. We study a variety of approaches to reduce the frequency of hallucinations, including data augmentation, dynamical systems and regularization techniques, showing that a data augmentation technique significantly reduces hallucination frequency. Finally, we analyze networks that produce hallucinations and show that there are signatures in the attention matrix as well as in the stability measures of the decoder. View details
    No Results Found