This is an internal, incomplete preview of a proposed change to the ACL Anthology.
For efficiency reasons, we generate only three BibTeX files per volume, and the preview may be incomplete in other ways, or contain mistakes.
Do not treat this content as an official publication.
Hyperparameter optimization (HPO) enhances neural machine translation (NMT) models but demands substantial computational resources. Successive halving, a multi-fidelity HPO method, mitigates this by early stopping unpromising models and allocating more resources to promising ones. This method is particularly relevant for NMT and large language models, which are computationally intensive. However, successive halving relies on a noisy estimation of model performance and assumes that early performance is highly correlated with final performance. We introduce a table lookup benchmark dataset to study the reliability of successive halving and propose best practices for its application in NMT and large language models.
Conversational speech translation is an important technology that fosters communication among people of different language backgrounds. Three-way parallel data in the form of source speech, source transcript, and target translation is usually required to train end-to-end systems. However, such datasets are not readily available and are expensive to create as this involves multiple annotation stages. In this paper, we investigate the use of synthetic data from generative models, namely machine translation and text-to-speech synthesis, for training conversational speech translation systems. We show that adding synthetic data to the training recipe increasingly improves end-to-end training performance, especially when limited real data is available. However, when no real data is available, no amount of synthetic data helps.
Recent evidence from cognitive science suggests that there exist two classes of cognitive representations within the spatial terms of a language, one represented geometrically (e.g., above, below) and the other functionally (e.g., on, in). It has been hypothesized that geometric terms are more constrained and are mastered relatively early in language learning, whereas functional terms are less constrained and are mastered over longer time periods (Landau, 2016). One consequence of this hypothesis is that these two classes should exhibit different cross-linguistic variability, which is supported by human elicitation studies. In this work we present to our knowledge the first corpus-based empirical test of this hypothesis. We develop a pipeline for extracting, isolating, and aligning spatial terms in basic locative constructions from parallel text. Using Shannon entropy to measure the variability of spatial term use across eight languages, we find supporting evidence that variability in functional terms differs significantly from that of geometric terms. We also perform latent variable modeling and find support for the division of spatial terms into geometric and functional classes.
Recent advances in automatic quality estimation for machine translation have exclusively focused on written language, leaving the speech modality underexplored. In this work, we formulate the task of quality estimation for speech translation (SpeechQE), construct a benchmark, and evaluate a family of systems based on cascaded and end-to-end architectures. In this process, we introduce a novel end-to-end system leveraging pre-trained text LLM. Results suggest that end-to-end approaches are better suited to estimating the quality of direct speech translation than using quality estimation systems designed for text in cascaded systems. More broadly, we argue that quality estimation of speech translation needs to be studied as a separate problem from that of text, and release our [data and models](https://github.com/h-j-han/SpeechQE) to guide further research in this space.
Zero-shot In-context learning is the phenomenon where models can perform a task given only the instructions. However, pre-trained large language models are known to be poorly calibrated for zero-shot tasks. One of the most effective approaches to handling this bias is to adopt a contrastive decoding objective, which accounts for the prior probability of generating the next token by conditioning on a context. This work introduces an Anti-Language Model objective with a decay factor designed to address the weaknesses of In-context Machine Translation. We conduct our experiments across 3 model types and sizes, 3 language directions, and for both greedy decoding and beam search. The proposed method outperforms other state-of-the-art decoding objectives, with up to 20 BLEU point improvement from the default objective in some settings.
Multilingual machine translation has proven immensely useful for both parameter efficiency and overall performance across many language pairs via complete multilingual parameter sharing. However, some language pairs in multilingual models can see worse performance than in bilingual models, especially in the one-to-many translation setting. Motivated by their empirical differences, we examine the geometric differences in representations from bilingual models versus those from one-to-many multilingual models. Specifically, we compute the isotropy of these representations using intrinsic dimensionality and IsoScore, in order to measure how the representations utilize the dimensions in their underlying vector space. Using the same evaluation data in both models, we find that for a given language pair, its multilingual model decoder representations are consistently less isotropic and occupy fewer dimensions than comparable bilingual model decoder representations. Additionally, we show that much of the anisotropy in multilingual decoder representations can be attributed to modeling language-specific information, therefore limiting remaining representational capacity.
Hyperparameter optimization is an important but often overlooked process in the research of deep learning technologies. To obtain a good model, one must carefully tune hyperparameters that determine the architecture and training algorithm. Insufficient tuning may result in poor results, while inequitable tuning may lead to exaggerated differences between models. We present a hyperparameter optimization toolkit for neural machine translation (NMT) to help researchers focus their time on the creative rather than the mundane. The toolkit is implemented as a wrapper on top of the open-source Sockeye NMT software. Using the Asynchronous Successive Halving Algorithm (ASHA), we demonstrate that it is possible to discover near-optimal models under a computational budget with little effort. Code: https://github.com/kevinduh/sockeye-recipes3Video demo: https://cs.jhu.edu/kevinduh/j/demo.mp4
Automated Machine Learning (AutoML) is an emerging field that has potential to impact how we build models in NLP. As an umbrella term that includes topics like hyperparameter optimization and neural architecture search, AutoML has recently become mainstream at major conferences such as NeurIPS, ICML, and ICLR. What does this mean to NLP? Currently, models are often built in an ad hoc process: we might borrow default hyperparameters from previous work and try a few variant architectures, but it is never guaranteed that final trained model is optimal. Automation can introduce rigor in this model-building process. This tutorial will summarize the main AutoML techniques and illustrate how to apply them to improve the NLP model-building process.
Back-translation is a data augmentation technique that has been shown to improve model quality through the creation of synthetic training bitext. Early studies showed the promise of the technique and follow on studies have produced additional refinements. We have undertaken a broad investigation using back-translation to train models from 60 languages into English; the majority of these languages are considered moderate- or low-resource languages. We observed consistent gains, though compared to prior work we saw conspicuous gains in quite a number of lower-resourced languages. We analyzed differences in translations between baseline and back-translation models, and observed many indications of improved translation quality. Translation of both rare and common terms is improved, and these improvements occur despite the less natural synthetic source-language text used in training.
The majority of existing work on sign language recognition encodes signed videos without explicitly acknowledging the phonological attributes of signs. Given that handshape is a vital parameter in sign languages, we explore the potential of handshape-aware sign language recognition. We augment the PHOENIX14T dataset with gloss-level handshape labels, resulting in the new PHOENIX14T-HS dataset. Two unique methods are proposed for handshape-inclusive sign language recognition: a single-encoder network and a dual-encoder network, complemented by a training strategy that simultaneously optimizes both the CTC loss and frame-level cross-entropy loss. The proposed methodology consistently outperforms the baseline performance. The dataset and code can be accessed at: www.anonymous.com.
This paper reports on the shared tasks organized by the 20th IWSLT Conference. The shared tasks address 9 scientific challenges in spoken language translation: simultaneous and offline translation, automatic subtitling and dubbing, speech-to-speech translation, multilingual, dialect and low-resource speech translation, and formality control. The shared tasks attracted a total of 38 submissions by 31 teams. The growing interest towards spoken language translation is also witnessed by the constantly increasing number of shared task organizers and contributors to the overview paper, almost evenly distributed across industry and academia.
We present a simple yet efficient method to enhance the quality of machine translation models trained on multimodal corpora by augmenting the training text with labels of detected objects in the corresponding video segments. We then test the effects of label augmentation in both baseline and two automatic speech recognition (ASR) conditions. In contrast with multimodal techniques that merge visual and textual features, our modular method is easy to implement and the results are more interpretable. Comparisons are made with Transformer translation architectures trained with baseline and augmented labels, showing improvements of up to +1.0 BLEU on the How2 dataset.
The phenomena of in-context learning has typically been thought of as “learning from examples”. In this work which focuses on Machine Translation, we present a perspective of in-context learning as the desired generation task maintaining coherency with its context, i.e., the prompt examples. We first investigate randomly sampled prompts across 4 domains, and find that translation performance improves when shown in-domain prompts. Next, we investigate coherency for the in-domain setting, which uses prompt examples from a moving window. We study this with respect to other factors that have previously been identified in the literature such as length, surface similarity and sentence embedding similarity. Our results across 3 models (GPTNeo2.7B, Bloom3B, XGLM2.9B), and three translation directions (en→{pt, de, fr}) suggest that the long-term coherency of the prompts and the test sentence is a good indicator of downstream translation performance. In doing so, we demonstrate the efficacy of in-context Machine Translation for on-the-fly adaptation.
While large language models have made remarkable advancements in natural language generation, their potential in machine translation, especially when fine-tuned, remains under-explored. In our study, we conduct comprehensive experiments, evaluating 15 publicly available language models on machine translation tasks. We compare the performance across three methodologies: zero-shot prompting, few-shot learning, and fine-tuning. Central to our approach is the use of QLoRA, an efficient fine-tuning method. On French-English, QLoRA fine-tuning outperforms both few-shot learning and models trained from scratch. This superiority is highlighted in both sentence-level and document-level translations, with a significant BLEU score improvement of 28.93 over the prompting method. Impressively, with QLoRA, the enhanced performance is achieved by fine-tuning a mere 0.77% of the model’s parameters.
Machine translation traditionally refers to translating from a single source language into a single target language. In recent years, the field has moved towards large neural models either translating from or into many languages. The model must be correctly cued to translate into the correct target language. This is typically done by prefixing language tokens onto the source or target sequence. The location and content of the prefix can vary and many use different approaches without much justification towards one approach or another. As a guidance to future researchers and directions for future work, we present a series of experiments that show how the positioning and type of a target language prefix token effects translation performance. We show that source side prefixes improve performance. Further, we find that the best language information to denote via tokens depends on the supported language set.
Pretrained multilingual sequence-to-sequence models have been successful in improving translation performance for mid- and lower-resourced languages. However, it is unclear if these models are helpful in the domain adaptation setting, and if so, how to best adapt them to both the domain and translation language pair. Therefore, in this work, we propose two major fine-tuning strategies: our language-first approach first learns the translation language pair via general bitext, followed by the domain via in-domain bitext, and our domain-first approach first learns the domain via multilingual in-domain bitext, followed by the language pair via language pair-specific in-domain bitext. We test our approach on 3 domains at different levels of data availability, and 5 language pairs. We find that models using an mBART initialization generally outperform those using a random Transformer initialization. This holds for languages even outside of mBART’s pretraining set, and can result in improvements of over +10 BLEU. Additionally, we find that via our domain-first approach, fine-tuning across multilingual in-domain corpora can lead to stark improvements in domain adaptation without sourcing additional out-of-domain bitext. In larger domain availability settings, our domain-first approach can be competitive with our language-first approach, even when using over 50X less data.
Very large language models have been shown to translate with few-shot in-context examples. However, they have not achieved state-of-art results for translating out of English. In this work, we investigate an extremely lightweight fixed-parameter method for conditioning a large language model to better translate into the target language. Our method introduces additional embeddings, known as prefix embeddings which do not interfere with the existing weights of the model. Using unsupervised and weakly semi-supervised methods that train only 0.0001% of the model parameters, the simple method improves ~0.2-1.3 BLEU points across 3 domains and 3 languages. We analyze the resulting embeddings’ training dynamics, and where they lie in the embedding space, and show that our trained embeddings can be used for both in-context translation, and diverse generation of the target sentence.
Hyperparameter tuning is important for achieving high accuracy in deep learning models, yet little interpretability work has focused on hyperparameters. We propose to use the Explainable Boosting Machine (EBM), a glassbox method, as a post-hoc analysis tool for understanding how hyperparameters influence model accuracy. We present a case study on Transformer models in machine translation to illustrate the kinds of insights that may be gleaned, and perform extensive analysis to test the robustness of EBM under different data conditions.
Bilingual lexicons form a critical component of various natural language processing applications, including unsupervised and semisupervised machine translation and crosslingual information retrieval. In this work, we improve bilingual lexicon induction performance across 40 language pairs with a graph-matching method based on optimal transport. The method is especially strong with low amounts of supervision.
The ability to extract high-quality translation dictionaries from monolingual word embedding spaces depends critically on the geometric similarity of the spaces—their degree of “isomorphism.” We address the root-cause of faulty cross-lingual mapping: that word embedding training resulted in the underlying spaces being non-isomorphic. We incorporate global measures of isomorphism directly into the skipgram loss function, successfully increasing the relative isomorphism of trained word embedding spaces and improving their ability to be mapped to a shared cross-lingual space. The result is improved bilingual lexicon induction in general data conditions, under domain mismatch, and with training algorithm dissimilarities. We release IsoVec at https://github.com/kellymarchisio/isovec.
Language diversity in NLP is critical in enabling the development of tools for a wide range of users.However, there are limited resources for building such tools for many languages, particularly those spoken in Africa.For search, most existing datasets feature few or no African languages, directly impacting researchers’ ability to build and improve information access capabilities in those languages.Motivated by this, we created AfriCLIRMatrix, a test collection for cross-lingual information retrieval research in 15 diverse African languages.In total, our dataset contains 6 million queries in English and 23 million relevance judgments automatically mined from Wikipedia inter-language links, covering many more African languages than any existing information retrieval test collection.In addition, we release BM25, dense retrieval, and sparse–dense hybrid baselines to provide a starting point for the development of future systems.We hope that these efforts can spur additional work in search for African languages.AfriCLIRMatrix can be downloaded at https://github.com/castorini/africlirmatrix.
In contexts where debate and deliberation are the norm, the participants are regularly presented with new information that conflicts with their original beliefs. When required to update their beliefs (belief alignment), they may choose arguments that align with their worldview (confirmation bias). We test this and competing hypotheses in a constraint-based modeling approach to predict the winning arguments in multi-party interactions in the Reddit Change My View and Intelligence Squared debates datasets. We adopt a hierarchical generative Variational Autoencoder as our model and impose structural constraints that reflect competing hypotheses about the nature of argumentation. Our findings suggest that in most settings, predictive models that anticipate winning arguments to be further from the initial argument of the opinion holder are more likely to succeed.
The evaluation campaign of the 19th International Conference on Spoken Language Translation featured eight shared tasks: (i) Simultaneous speech translation, (ii) Offline speech translation, (iii) Speech to speech translation, (iv) Low-resource speech translation, (v) Multilingual speech translation, (vi) Dialect speech translation, (vii) Formality control for speech translation, (viii) Isometric speech translation. A total of 27 teams participated in at least one of the shared tasks. This paper details, for each shared task, the purpose of the task, the data that were released, the evaluation metrics that were applied, the submissions that were received and the results that were achieved.
Translation of the noisy, informal language found in social media has been an understudied problem, with a principal factor being the limited availability of translation corpora in many languages. To address this need we have developed a new corpus containing over 200,000 translations of microblog posts that supports translation of thirteen languages into English. The languages are: Arabic, Chinese, Farsi, French, German, Hindi, Korean, Pashto, Portuguese, Russian, Spanish, Tagalog, and Urdu. We are releasing these data as the Multilingual Microblog Translation Corpus to support futher research in translation of informal language. We establish baselines using this new resource, and we further demonstrate the utility of the corpus by conducting experiments with fine-tuning to improve translation quality from a high performing neural machine translation (NMT) system. Fine-tuning provided substantial gains, ranging from +3.4 to +11.1 BLEU. On average, a relative gain of 21% was observed, demonstrating the utility of the corpus.
“Transcription bottlenecks”, created by a shortage of effective human transcribers (i.e., transcriber shortage), are one of the main challenges to endangered language (EL) documentation. Automatic speech recognition (ASR) has been suggested as a tool to overcome such bottlenecks. Following this suggestion, we investigated the effectiveness for EL documentation of end-to-end ASR, which unlike Hidden Markov Model ASR systems, eschews linguistic resources but is instead more dependent on large-data settings. We open source a Yoloxóchitl Mixtec EL corpus. First, we review our method in building an end-to-end ASR system in a way that would be reproducible by the ASR community. We then propose a novice transcription correction task and demonstrate how ASR systems and novice transcribers can work together to improve EL documentation. We believe this combinatory methodology would mitigate the transcription bottleneck and transcriber shortage that hinders EL documentation.
Probabilistic topic models in low data resource scenarios are faced with less reliable estimates due to sparsity of discrete word co-occurrence counts, and do not have the luxury of retraining word or topic embeddings using neural methods. In this challenging resource constrained setting, we explore mixture models which interpolate between the discrete and continuous topic-word distributions that utilise pre-trained embeddings to improve topic coherence. We introduce an automatic trade-off between the discrete and continuous representations via an adaptive mixture coefficient, which places greater weight on the discrete representation when the corpus statistics are more reliable. The adaptive mixture coefficient takes into account global corpus statistics, and the uncertainty in each topic’s continuous distributions. Our approach outperforms the fully discrete, fully continuous, and static mixture model on topic coherence in low resource settings. We additionally demonstrate the generalisability of our method by extending it to handle multilingual document collections.
We observe that the development cross-entropy loss of supervised neural machine translation models scales like a power law with the amount of training data and the number of non-embedding parameters in the model. We discuss some practical implications of these results, such as predicting BLEU achieved by large scale models and predicting the ROI of labeling data in low-resource language pairs.
Much recent work in bilingual lexicon induction (BLI) views word embeddings as vectors in Euclidean space. As such, BLI is typically solved by finding a linear transformation that maps embeddings to a common space. Alternatively, word embeddings may be understood as nodes in a weighted graph. This framing allows us to examine a node’s graph neighborhood without assuming a linear transform, and exploits new techniques from the graph matching optimization literature. These contrasting approaches have not been compared in BLI so far. In this work, we study the behavior of Euclidean versus graph-based approaches to BLI under differing data conditions and show that they complement each other when combined. We release our code at https://github.com/kellymarchisio/euc-v-graph-bli.
Successful Machine Translation (MT) deployment requires understanding not only the intrinsic qualities of MT output, such as fluency and adequacy, but also user perceptions. Users who do not understand the source language respond to MT output based on their perception of the likelihood that the meaning of the MT output matches the meaning of the source text. We refer to this as believability. Output that is not believable may be off-putting to users, but believable MT output with incorrect meaning may mislead them. In this work, we study the relationship of believability to fluency and adequacy by applying traditional MT direct assessment protocols to annotate all three features on the output of neural MT systems. Quantitative analysis of these annotations shows that believability is closely related to but distinct from fluency, and initial qualitative analysis suggests that semantic features may account for the difference.
This paper describes the ESPnet-ST group’s IWSLT 2021 submission in the offline speech translation track. This year we made various efforts on training data, architecture, and audio segmentation. On the data side, we investigated sequence-level knowledge distillation (SeqKD) for end-to-end (E2E) speech translation. Specifically, we used multi-referenced SeqKD from multiple teachers trained on different amounts of bitext. On the architecture side, we adopted the Conformer encoder and the Multi-Decoder architecture, which equips dedicated decoders for speech recognition and translation tasks in a unified encoder-decoder model and enables search in both source and target language spaces during inference. We also significantly improved audio segmentation by using the pyannote.audio toolkit and merging multiple short segments for long context modeling. Experimental evaluations showed that each of them contributed to large improvements in translation performance. Our best E2E system combined all the above techniques with model ensembling and achieved 31.4 BLEU on the 2-ref of tst2021 and 21.2 BLEU and 19.3 BLEU on the two single references of tst2021.
In supervised learning, a well-trained model should be able to recover ground truth accurately, i.e. the predicted labels are expected to resemble the ground truth labels as much as possible. Inspired by this, we formulate a difficulty criterion based on the recovery degrees of training examples. Motivated by the intuition that after skimming through the training corpus, the neural machine translation (NMT) model “knows” how to schedule a suitable curriculum according to learning difficulty, we propose a self-guided curriculum learning strategy that encourages the NMT model to learn from easy to hard on the basis of recovery degrees. Specifically, we adopt sentence-level BLEU score as the proxy of recovery degree. Experimental results on translation benchmarks including WMT14 English-German and WMT17 Chinese-English demonstrate that our proposed method considerably improves the recovery degree, thus consistently improving the translation performance.
A cascaded Sign Language Translation system first maps sign videos to gloss annotations and then translates glosses into a spoken languages. This work focuses on the second-stage gloss translation component, which is challenging due to the scarcity of publicly available parallel data. We approach gloss translation as a low-resource machine translation task and investigate two popular methods for improving translation quality: hyperparameter search and backtranslation. We discuss the potentials and pitfalls of these methods based on experiments on the RWTH-PHOENIX-Weather 2014T dataset.
We present CLIReval, an easy-to-use toolkit for evaluating machine translation (MT) with the proxy task of cross-lingual information retrieval (CLIR). Contrary to what the project name might suggest, CLIReval does not actually require any annotated CLIR dataset. Instead, it automatically transforms translations and references used in MT evaluations into a synthetic CLIR dataset; it then sets up a standard search engine (Elasticsearch) and computes various information retrieval metrics (e.g., mean average precision) by treating the translations as documents to be retrieved. The idea is to gauge the quality of MT by its impact on the document translation approach to CLIR. As a case study, we run CLIReval on the “metrics shared task” of WMT2019; while this extrinsic metric is not intended to replace popular intrinsic metrics such as BLEU, results suggest CLIReval is competitive in many language pairs in terms of correlation to human judgments of quality. CLIReval is publicly available at https://github.com/ssun32/CLIReval.
We present ESPnet-ST, which is designed for the quick development of speech-to-speech translation systems in a single framework. ESPnet-ST is a new project inside end-to-end speech processing toolkit, ESPnet, which integrates or newly implements automatic speech recognition, machine translation, and text-to-speech functions for speech translation. We provide all-in-one recipes including data pre-processing, feature extraction, training, and decoding pipelines for a wide range of benchmark datasets. Our reproducible results can match or even outperform the current state-of-the-art performances; these pre-trained models are downloadable. The toolkit is publicly available at https://github.com/espnet/espnet.
We present CLIRMatrix, a massively large collection of bilingual and multilingual datasets for Cross-Lingual Information Retrieval extracted automatically from Wikipedia. CLIRMatrix comprises (1) BI-139, a bilingual dataset of queries in one language matched with relevant documents in another language for 139x138=19,182 language pairs, and (2) MULTI-8, a multilingual dataset of queries and documents jointly aligned in 8 different languages. In total, we mined 49 million unique queries and 34 billion (query, document, label) triplets, making it the largest and most comprehensive CLIR dataset to date. This collection is intended to support research in end-to-end neural information retrieval and is publicly available at [url]. We provide baseline neural model results on BI-139, and evaluate MULTI-8 in both single-language retrieval and mix-language retrieval settings.
Research in machine translation (MT) is developing at a rapid pace. However, most work in the community has focused on languages where large amounts of digital resources are available. In this study, we benchmark state of the art statistical and neural machine translation systems on two African languages which do not have large amounts of resources: Somali and Swahili. These languages are of social importance and serve as test-beds for developing technologies that perform reasonably well despite the low-resource constraint. Our findings suggest that statistical machine translation (SMT) and neural machine translation (NMT) can perform similarly in low-resource scenarios, but neural systems require more careful tuning to match performance. We also investigate how to exploit additional data, such as bilingual text harvested from the web, or user dictionaries; we find that NMT can significantly improve in performance with the use of these additional data. Finally, we survey the landscape of machine translation resources for the languages of Africa and provide some suggestions for promising future research directions.
We explore best practices for training small, memory efficient machine translation models with sequence-level knowledge distillation in the domain adaptation setting. While both domain adaptation and knowledge distillation are widely-used, their interaction remains little understood. Our large-scale empirical results in machine translation (on three language pairs with three domains each) suggest distilling twice for best performance: once using general-domain data and again using in-domain data with an adapted teacher.
Pre-trained universal feature extractors, such as BERT for natural language processing and VGG for computer vision, have become effective methods for improving deep learning models without requiring more labeled data. While effective, feature extractors like BERT may be prohibitively large for some deployment scenarios. We explore weight pruning for BERT and ask: how does compression during pre-training affect transfer learning? We find that pruning affects transfer learning in three broad regimes. Low levels of pruning (30-40%) do not affect pre-training loss or transfer to downstream tasks at all. Medium levels of pruning increase the pre-training loss and prevent useful pre-training information from being transferred to downstream tasks. High levels of pruning additionally prevent models from fitting downstream datasets, leading to further degradation. Finally, we observe that fine-tuning BERT on a specific task does not improve its prunability. We conclude that BERT can be pruned once during pre-training rather than separately for each task without affecting performance.
Data privacy is an important issue for “machine learning as a service” providers. We focus on the problem of membership inference attacks: Given a data sample and black-box access to a model’s API, determine whether the sample existed in the model’s training data. Our contribution is an investigation of this problem in the context of sequence-to-sequence models, which are important in applications such as machine translation and video captioning. We define the membership inference problem for sequence generation, provide an open dataset based on state-of-the-art machine translation models, and report initial results on whether these models leak private information against several kinds of membership inference attacks.
Hyperparameter selection is a crucial part of building neural machine translation (NMT) systems across both academia and industry. Fine-grained adjustments to a model’s architecture or training recipe can mean the difference between a positive and negative research result or between a state-of-the-art and underperforming system. While recent literature has proposed methods for automatic hyperparameter optimization (HPO), there has been limited work on applying these methods to neural machine translation (NMT), due in part to the high costs associated with experiments that train large numbers of model variants. To facilitate research in this space, we introduce a lookup-based approach that uses a library of pre-trained models for fast, low cost HPO experimentation. Our contributions include (1) the release of a large collection of trained NMT models covering a wide range of hyperparameters, (2) the proposal of targeted metrics for evaluating HPO methods on NMT, and (3) a reproducible benchmark of several HPO methods against our model library, including novel graph-based and multiobjective methods.
Despite the reported success of unsupervised machine translation (MT), the field has yet to examine the conditions under which the methods succeed and fail. We conduct an extensive empirical evaluation using dissimilar language pairs, dissimilar domains, and diverse datasets. We find that performance rapidly deteriorates when source and target corpora are from different domains, and that stochasticity during embedding training can dramatically affect downstream results. We additionally find that unsupervised MT performance declines when source and target languages use different scripts, and observe very poor performance on authentic low-resource language pairs. We advocate for extensive empirical evaluation of unsupervised MT systems to highlight failure points and encourage continued research on the most promising paradigms. We release our preprocessed dataset to encourage evaluations that stress-test systems under multiple data conditions.
This paper describes the ESPnet submissions to the How2 Speech Translation task at IWSLT2019. In this year, we mainly build our systems based on Transformer architectures in all tasks and focus on the end-to-end speech translation (E2E-ST). We first compare RNN-based models and Transformer, and then confirm Transformer models significantly and consistently outperform RNN models in all tasks and corpora. Next, we investigate pre-training of E2E-ST models with the ASR and MT tasks. On top of the pre-training, we further explore knowledge distillation from the NMT model and the deeper speech encoder, and confirm drastic improvements over the baseline model. All of our codes are publicly available in ESPnet.
Bilingual lexicons are valuable resources used by professional human translators. While these resources can be easily incorporated in statistical machine translation, it is unclear how to best do so in the neural framework. In this work, we present the HABLex dataset, designed to test methods for bilingual lexicon integration into neural machine translation. Our data consists of human generated alignments of words and phrases in machine translation test sets in three language pairs (Russian-English, Chinese-English, and Korean-English), resulting in clean bilingual lexicons which are well matched to the reference. We also present two simple baselines - constrained decoding and continued training - and an improvement to continued training to address overfitting.
We unify different broad-coverage semantic parsing tasks into a transduction parsing paradigm, and propose an attention-based neural transducer that incrementally builds meaning representation via a sequence of semantic relations. By leveraging multiple attention mechanisms, the neural transducer can be effectively trained without relying on a pre-trained aligner. Experiments separately conducted on three broad-coverage semantic parsing tasks – AMR, SDP and UCCA – demonstrate that our attention-based neural transducer improves the state of the art on both AMR and UCCA, and is competitive with the state of the art on SDP.
We introduce a curriculum learning approach to adapt generic neural machine translation models to a specific domain. Samples are grouped by their similarities to the domain of interest and each group is fed to the training algorithm with a particular schedule. This approach is simple to implement on top of any neural framework or architecture, and consistently outperforms both unadapted and adapted baselines in experiments with two distinct domains and two language pairs.
Continued training is an effective method for domain adaptation in neural machine translation. However, in-domain gains from adaptation come at the expense of general-domain performance. In this work, we interpret the drop in general-domain performance as catastrophic forgetting of general-domain knowledge. To mitigate it, we adapt Elastic Weight Consolidation (EWC)—a machine learning method for learning a new task without forgetting previous tasks. Our method retains the majority of general-domain performance lost in continued training without degrading in-domain performance, outperforming the previous state-of-the-art. We also explore the full range of general-domain performance available when some in-domain degradation is acceptable.
We propose an attention-based model that treats AMR parsing as sequence-to-graph transduction. Unlike most AMR parsers that rely on pre-trained aligners, external semantic resources, or data augmentation, our proposed parser is aligner-free, and it can be effectively trained with limited amounts of labeled AMR data. Our experimental results outperform all previously reported SMATCH scores, on both AMR 2.0 (76.3% on LDC2017T10) and AMR 1.0 (70.2% on LDC2014T12).
When translating diglossic languages such as Arabic, situations may arise where we would like to translate a text but do not know which dialect it is. A traditional approach to this problem is to design dialect identification systems and dialect-specific machine translation systems. However, under the recent paradigm of neural machine translation, shared multi-dialectal systems have become a natural alternative. Here we explore under which conditions it is beneficial to perform dialect identification for Arabic neural machine translation versus using a general system for all dialects.
Our submission to the MADAR shared task on Arabic dialect identification employed a language modeling technique called Prediction by Partial Matching, an ensemble of neural architectures, and sources of additional data for training word embeddings and auxiliary language models. We found several of these techniques provided small boosts in performance, though a simple character-level language model was a strong baseline, and a lower-order LM achieved best performance on Subtask 2. Interestingly, word embeddings provided no consistent benefit, and ensembling struggled to outperform the best component submodel. This suggests the variety of architectures are learning redundant information, and future work may focus on encouraging decorrelated learning.
We describe the JHU submissions to the French–English, Japanese–English, and English–Japanese Robustness Task at WMT 2019. Our goal was to evaluate the performance of baseline systems on both the official noisy test set as well as news data, in order to ensure that performance gains in the latter did not come at the expense of general-domain performance. To this end, we built straightforward 6-layer Transformer models and experimented with a handful of variables including subword processing (FR→EN) and a handful of hyperparameters settings (JA↔EN). As expected, our systems performed reasonably.
This paper describes the Johns Hopkins University (JHU) and Kyoto University submissions to the Speech Translation evaluation campaign at IWSLT2018. Our end-to-end speech translation systems are based on ESPnet and implements an attention-based encoder-decoder model. As comparison, we also experiment with a pipeline system that uses independent neural network systems for both the speech transcription and text translation components. We find that a transfer learning approach that bootstraps the end-to-end speech translation system with speech transcription system’s parameters is important for training on small datasets.
We introduce the task of cross-lingual decompositional semantic parsing: mapping content provided in a source language into a decompositional semantic analysis based on a target language. We present: (1) a form of decompositional semantic analysis designed to allow systems to target varying levels of structural complexity (shallow to deep analysis), (2) an evaluation metric to measure the similarity between system output and reference semantic analysis, (3) an end-to-end model with a novel annotating mechanism that supports intra-sentential coreference, and (4) an evaluation dataset on which our model outperforms strong baselines by at least 1.75 F1 score.
Cross-lingual information retrieval (CLIR) is a document retrieval task where the documents are written in a language different from that of the user’s query. This is a challenging problem for data-driven approaches due to the general lack of labeled training data. We introduce a large-scale dataset derived from Wikipedia to support CLIR research in 25 languages. Further, we present a simple yet effective neural learning-to-rank model that shares representations across languages and reduces the data requirement. This model can exploit training data in, for example, Japanese-English CLIR to improve the results of Swahili-English CLIR.
We propose a simple yet robust stochastic answer network (SAN) that simulates multi-step reasoning in machine reading comprehension. Compared to previous work such as ReasoNet which used reinforcement learning to determine the number of steps, the unique feature is the use of a kind of stochastic prediction dropout on the answer module (final layer) of the neural network during the training. We show that this simple trick improves robustness and achieves results competitive to the state-of-the-art on the Stanford Question Answering Dataset (SQuAD), the Adversarial SQuAD, and the Microsoft MAchine Reading COmprehension Dataset (MS MARCO).
Cross-lingual information extraction (CLIE) is an important and challenging task, especially in low resource scenarios. To tackle this challenge, we propose a training method, called Halo, which enforces the local region of each hidden state of a neural model to only generate target tokens with the same semantic structure tag. This simple but powerful technique enables a neural model to learn semantics-aware representations that are robust to noise, without introducing any extra parameter, thus yielding better generalization in both high and low resource settings.
Fine-grained entity typing is the task of assigning fine-grained semantic types to entity mentions. We propose a neural architecture which learns a distributional semantic representation that leverages a greater amount of semantic context – both document and sentence level information – than prior work. We find that additional context improves performance, with further improvements gained by utilizing adaptive classification thresholds. Experiments show that our approach without reliance on hand-crafted features achieves the state-of-the-art results on three benchmark datasets.
Neural machine translation has achieved impressive results in the last few years, but its success has been limited to settings with large amounts of parallel data. One way to improve NMT for lower-resource settings is to initialize a word-based NMT model with pretrained word embeddings. However, rare words still suffer from lower quality word embeddings when trained with standard word-level objectives. We introduce word embeddings that utilize morphological resources, and compare to purely unsupervised alternatives. We work with Arabic, a morphologically rich language with available linguistic resources, and perform Ar-to-En MT experiments on a small corpus of TED subtitles. We find that word embeddings utilizing subword information consistently outperform standard word embeddings on a word similarity task and as initialization of the source word embeddings in a low-resource NMT system.
Supervised domain adaptation—where a large generic corpus and a smaller in-domain corpus are both available for training—is a challenge for neural machine translation (NMT). Standard practice is to train a generic model and use it to initialize a second model, then continue training the second model on in-domain data to produce an in-domain model. We add an auxiliary term to the training objective during continued training that minimizes the cross entropy between the in-domain model’s output word distribution and that of the out-of-domain model to prevent the model’s output from differing too much from the original out-of-domain model. We perform experiments on EMEA (descriptions of medicines) and TED (rehearsed presentations), initialized from a general domain (WMT) model. Our method shows improvements over standard continued training by up to 1.5 BLEU.
To better understand the effectiveness of continued training, we analyze the major components of a neural machine translation system (the encoder, decoder, and each embedding space) and consider each component’s contribution to, and capacity for, domain adaptation. We find that freezing any single component during continued training has minimal impact on performance, and that performance is surprisingly good when a single component is adapted while holding the rest of the model fixed. We also find that continued training does not move the model very far from the out-of-domain model, compared to a sensitivity analysis metric, suggesting that the out-of-domain model can provide a good generic initialization for the new domain.
We report on the efforts of the Johns Hopkins University to develop neural machine translation systems for the shared task for news translation organized around the Conference for Machine Translation (WMT) 2018. We developed systems for German–English, English– German, and Russian–English. Our novel contributions are iterative back-translation and fine-tuning on test sets from prior years.
Neural machine translation (NMT) systems have demonstrated promising results in recent years. However, non-trivial amounts of manual effort are required for tuning network architectures, training configurations, and pre-processing settings such as byte pair encoding (BPE). In this study, we propose an evolution strategy based automatic tuning method for NMT. In particular, we apply the covariance matrix adaptation-evolution strategy (CMA-ES), and investigate a Pareto-based multi-objective CMA-ES to optimize the translation performance and computational time jointly. Experimental results show that the proposed method automatically finds NMT systems that outperform the initial manual setting.
Cross-lingual information extraction is the task of distilling facts from foreign language (e.g. Chinese text) into representations in another language that is preferred by the user (e.g. English tuples). Conventional pipeline solutions decompose the task as machine translation followed by information extraction (or vice versa). We propose a joint solution with a neural sequence model, and show that it outperforms the pipeline in a cross-lingual open information extraction setting by 1-4 BLEU and 0.5-0.8 F1.
Cross-lingual open information extraction is the task of distilling facts from the source language into representations in the target language. We propose a novel encoder-decoder model for this problem. It employs a novel selective decoding mechanism, which explicitly models the sequence labeling process as well as the sequence generation process on the decoder side. Compared to a standard encoder-decoder model, selective decoding significantly increases the performance on a Chinese-English cross-lingual open IE dataset by 3.87-4.49 BLEU and 1.91-5.92 F1. We also extend our approach to low-resource scenarios, and gain promising improvement.
Reading comprehension (RC) is a challenging task that requires synthesis of information across sentences and multiple turns of reasoning. Using a state-of-the-art RC model, we empirically investigate the performance of single-turn and multiple-turn reasoning on the SQuAD and MS MARCO datasets. The RC model is an end-to-end neural network with iterative attention, and uses reinforcement learning to dynamically control the number of turns. We find that multiple-turn reasoning outperforms single-turn reasoning for all question and answer types; further, we observe that enabling a flexible number of turns generally improves upon a fixed multiple-turn strategy. %across all question types, and is particularly beneficial to questions with lengthy, descriptive answers. We achieve results competitive to the state-of-the-art on these two datasets.
We propose to unify a variety of existing semantic classification tasks, such as semantic role labeling, anaphora resolution, and paraphrase detection, under the heading of Recognizing Textual Entailment (RTE). We present a general strategy to automatically generate one or more sentential hypotheses based on an input sentence and pre-existing manual semantic annotations. The resulting suite of datasets enables us to probe a statistical RTE model’s performance on different aspects of semantics. We demonstrate the value of this approach by investigating the behavior of a popular neural network RTE model.
Domain adaptation is a major challenge for neural machine translation (NMT). Given unknown words or new domains, NMT systems tend to generate fluent translations at the expense of adequacy. We present a stack-based lattice search algorithm for NMT and show that constraining its search space with lattices generated by phrase-based machine translation (PBMT) improves robustness. We report consistent BLEU score gains across four diverse domain adaptation tasks involving medical, IT, Koran, or subtitles texts.
Low-resource named entity recognition is still an open problem in NLP. Most state-of-the-art systems require tens of thousands of annotated sentences in order to obtain high performance. However, for most of the world’s languages it is unfeasible to obtain such annotation. In this paper, we present a transfer learning scheme, whereby we train character-level neural CRFs to predict named entities for both high-resource languages and low-resource languages jointly. Learning character representations for multiple related languages allows knowledge transfer from the high-resource languages to the low-resource ones, improving F1 by up to 9.8 points.
We show how to adapt bilingual word embeddings (BWE’s) to bootstrap a cross-lingual name-entity recognition (NER) system in a language with no labeled data. We assume a setting where we are given a comparable corpus with NER labels for the source language only; our goal is to build a NER model for the target language. The proposed multi-task model jointly trains bilingual word embeddings while optimizing a NER objective. This creates word embeddings that are both shared between languages and fine-tuned for the NER task.
Computer Assisted Discovery Extraction and Translation (CADET) is a workbench for helping knowledge workers find, label, and translate documents of interest. It combines a multitude of analytics together with a flexible environment for customizing the workflow for different users. This open-source framework allows for easy development of new research prototypes using a micro-service architecture based atop Docker and Apache Thrift.
Humans have the capacity to draw common-sense inferences from natural language: various things that are likely but not certain to hold based on established discourse, and are rarely stated explicitly. We propose an evaluation of automated common-sense inference based on an extension of recognizing textual entailment: predicting ordinal human responses on the subjective likelihood of an inference holding in a given context. We describe a framework for extracting common-sense knowledge from corpora, which is then used to construct a dataset for this ordinal entailment task. We train a neural sequence-to-sequence model on this dataset, which we use to score and generate possible inferences. Further, we annotate subsets of previously established datasets via our ordinal annotation protocol in order to then analyze the distinctions between these and what we have constructed.
This paper presents NTT-NAIST SMT systems for English-German and German-English MT tasks of the IWSLT 2014 evaluation campaign. The systems are based on generalized minimum Bayes risk system combination of three SMT systems using the forest-to-string, syntactic preordering, and phrase-based translation formalisms. Individual systems employ training data selection for domain adaptation, truecasing, compound word splitting (for GermanEnglish), interpolated n-gram language models, and hypotheses rescoring using recurrent neural network language models.
Syntactic parsing is a fundamental natural language processing technology that has proven useful in machine translation, language modeling, sentence segmentation, and a number of other applications related to speech translation. However, there is a paucity of manually annotated syntactic parsing resources for speech, and particularly for the lecture speech that is the current target of the IWSLT translation campaign. In this work, we present a new manually annotated treebank of TED talks that we hope will prove useful for investigation into the interaction between syntax and these speechrelated applications. The first version of the corpus includes 1,217 sentences and 23,158 words manually annotated with parse trees, and aligned with translations in 26-43 different languages. In this paper we describe the collection of the corpus, and an analysis of its various characteristics.
Synthetic word analysis is a potentially important but relatively unexplored problem in Chinese natural language processing. Two issues with the conventional pipeline methods involving word segmentation are (1) the lack of a common segmentation standard and (2) the poor segmentation performance on OOV words. These issues may be circumvented if we adopt the view of character-based parsing, providing both internal structures to synthetic words and global structure to sentences in a seamless fashion. However, the accuracy of synthetic word parsing is not yet satisfactory, due to the lack of research. In view of this, we propose and present experiments on several synthetic word parsers. Additionally, we demonstrate the usefulness of incorporating large unlabelled corpora and a dictionary for this task. Our parsers significantly outperform the baseline (a pipeline method).
This paper presents NTT-NAIST SMT systems for English-German and German-English MT tasks of the IWSLT 2013 evaluation campaign. The systems are based on generalized minimum Bayes risk system combination of three SMT systems: forest-to-string, hierarchical phrase-based, phrasebased with pre-ordering. Individual SMT systems include data selection for domain adaptation, rescoring using recurrent neural net language models, interpolated language models, and compound word splitting (only for German-English).
This paper describes the NAIST statistical machine translation system for the IWSLT2012 Evaluation Campaign. We participated in all TED Talk tasks, for a total of 11 language-pairs. For all tasks, we use the Moses phrase-based decoder and its experiment management system as a common base for building translation systems. The focus of our work is on performing a comprehensive comparison of a multitude of existing techniques for the TED task, exploring issues such as out-of-domain data filtering, minimum Bayes risk decoding, MERT vs. PRO tuning, word alignment combination, and morphology.
This paper describes the University of Washington’s system for the 2009 International Workshop on Spoken Language Translation (IWSLT) evaluation campaign. Two systems were developed, one each for the BTEC Chinese-to-English and Arabic-to-English tracks. We describe experiments with different preprocessing and alignment combination schemes. Our main focus this year was on exploring a novel semi-supervised approach to N-best list reranking; however, this method yielded inconclusive results.