arXiv subjects

Matteo Negri

Publications and source records attributed to Matteo Negri.

At least 109 records · Page 6Linked to original sources

arXiv2020

Is 42 the Answer to Everything in Subtitling-oriented Speech Translation?

Subtitling is becoming increasingly important for disseminating information, given the enormous amounts of audiovisual content becoming available daily. Although Neural Machine Translation (NMT) can speed up the process of translating audiovisual content, large manual effort is still required for transcribing the source language, and for spotting and segmenting the text into proper subtitles. Creating proper subtitles in terms of timing and segmentation highly depends on information present in the audio (utterance duration, natural pauses). In this work, we explore two methods for applying Speech Translation (ST) to subtitling: a) a direct end-to-end and b) a classical cascade approach. We discuss the benefit of having access to the source language speech for improving the conformity of the generated subtitles to the spatial and temporal subtitling constraints and show that length is not the answer to everything in the case of subtitling-oriented ST.

cs.CL↗

arXiv2020

Low Resource Neural Machine Translation: A Benchmark for Five African Languages

Recent advents in Neural Machine Translation (NMT) have shown improvements in low-resource language (LRL) translation tasks. In this work, we benchmark NMT between English and five African LRL pairs (Swahili, Amharic, Tigrigna, Oromo, Somali [SATOS]). We collected the available resources on the SATOS languages to evaluate the current state of NMT for LRLs. Our evaluation, comparing a baseline single language pair NMT model against semi-supervised learning, transfer learning, and multilingual modeling, shows significant performance improvements both in the En-LRL and LRL-En directions. In terms of averaged BLEU score, the multilingual approach shows the largest gains, up to +5 points, in six out of ten translation directions. To demonstrate the generalization capability of each model, we also report results on multi-domain test sets. We release the standardized experimental data and the test sets for future works addressing the challenges of NMT in under-resourced settings, in particular for the SATOS languages.

cs.CL↗

arXiv2020

MuST-Cinema: a Speech-to-Subtitles corpus

Growing needs in localising audiovisual content in multiple languages through subtitles call for the development of automatic solutions for human subtitling. Neural Machine Translation (NMT) can contribute to the automatisation of subtitling, facilitating the work of human subtitlers and reducing turn-around times and related costs. NMT requires high-quality, large, task-specific training data. The existing subtitling corpora, however, are missing both alignments to the source language audio and important information about subtitle breaks. This poses a significant limitation for developing efficient automatic approaches for subtitling, since the length and form of a subtitle directly depends on the duration of the utterance. In this work, we present MuST-Cinema, a multilingual speech translation corpus built from TED subtitles. The corpus is comprised of (audio, transcription, translation) triplets. Subtitle breaks are preserved by inserting special symbols. We show that the corpus can be used to build models that efficiently segment sentences into subtitles and propose a method for annotating existing subtitling corpora with subtitle breaks, conforming to the constraint of length.

cs.CL↗

arXiv2019

Adapting Multilingual Neural Machine Translation to Unseen Languages

Multilingual Neural Machine Translation (MNMT) for low-resource languages (LRL) can be enhanced by the presence of related high-resource languages (HRL), but the relatedness of HRL usually relies on predefined linguistic assumptions about language similarity. Recently, adapting MNMT to a LRL has shown to greatly improve performance. In this work, we explore the problem of adapting an MNMT model to an unseen LRL using data selection and model adaptation. In order to improve NMT for LRL, we employ perplexity to select HRL data that are most similar to the LRL on the basis of language distance. We extensively explore data selection in popular multilingual NMT settings, namely in (zero-shot) translation, and in adaptation from a multilingual pre-trained model, for both directions (LRL-en). We further show that dynamic adaptation of the model's vocabulary results in a more favourable segmentation for the LRL in comparison with direct adaptation. Experiments show reductions in training time and significant performance gains over LRL baselines, even with zero LRL data (+13.0 BLEU), up to +17.0 BLEU for pre-trained multilingual model dynamic adaptation with related data selection. Our method outperforms current approaches, such as massively multilingual models and data augmentation, on four LRL.

cs.CL↗

arXiv2019

A quasi-static model for craquelure patterns

We consider the quasi-static evolution of a brittle layer on a stiff substrate; adhesion between layers is assumed to be elastic. Employing a phase-field approach we obtain the quasi-static evolution as the limit of time-discrete evolutions computed by an alternate minimization scheme. We study the limit evolution, providing a qualitative discussion of its behaviour and a rigorous characterization, in terms of parametrized balanced viscosity evolutions. Further, we study the transition layer of the phase-field, in a simplified setting, and show that it governs the spacing of cracks in the first stages of the evolution. Numerical results show a good consistency with the theoretical study and the local morphology of real life craquelure patterns.

math.AP↗

arXiv2019

Instance-Based Model Adaptation For Direct Speech Translation

Despite recent technology advancements, the effectiveness of neural approaches to end-to-end speech-to-text translation is still limited by the paucity of publicly available training corpora. We tackle this limitation with a method to improve data exploitation and boost the system's performance at inference time. Our approach allows us to customize "on the fly" an existing model to each incoming translation request. At its core, it exploits an instance selection procedure to retrieve, from a given pool of data, a small set of samples similar to the input query in terms of latent properties of its audio signal. The retrieved samples are then used for an instance-specific fine-tuning of the model. We evaluate our approach in three different scenarios. In all data conditions (different languages, in/out-of-domain adaptation), our instance-based adaptation yields coherent performance gains over static models.

cs.CL↗

arXiv2019

One-To-Many Multilingual End-to-end Speech Translation

Nowadays, training end-to-end neural models for spoken language translation (SLT) still has to confront with extreme data scarcity conditions. The existing SLT parallel corpora are indeed orders of magnitude smaller than those available for the closely related tasks of automatic speech recognition (ASR) and machine translation (MT), which usually comprise tens of millions of instances. To cope with data paucity, in this paper we explore the effectiveness of transfer learning in end-to-end SLT by presenting a multilingual approach to the task. Multilingual solutions are widely studied in MT and usually rely on ``\textit{target forcing}'', in which multilingual parallel data are combined to train a single model by prepending to the input sequences a language token that specifies the target language. However, when tested in speech translation, our experiments show that MT-like \textit{target forcing}, used as is, is not effective in discriminating among the target languages. Thus, we propose a variant that uses target-language embeddings to shift the input representations in different portions of the space according to the language, so to better support the production of output in the desired target language. Our experiments on end-to-end SLT from English into six languages show important improvements when translating into similar languages, especially when these are supported by scarce data. Further improvements are obtained when using English ASR data as an additional language (up to $+2.5$ BLEU points).

cs.CL↗

arXiv2019

Machine Translation for Machines: the Sentiment Classification Use Case

We propose a neural machine translation (NMT) approach that, instead of pursuing adequacy and fluency ("human-oriented" quality criteria), aims to generate translations that are best suited as input to a natural language processing component designed for a specific downstream task (a "machine-oriented" criterion). Towards this objective, we present a reinforcement learning technique based on a new candidate sampling strategy, which exploits the results obtained on the downstream task as weak feedback. Experiments in sentiment classification of Twitter data in German and Italian show that feeding an English classifier with machine-oriented translations significantly improves its performance. Classification results outperform those obtained with translations produced by general-purpose NMT models as well as by an approach based on reinforcement learning. Moreover, our results on both languages approximate the classification accuracy computed on gold standard English tweets.

cs.CL↗

arXiv2019

Natural representation of composite data with replicated autoencoders

Generative processes in biology and other fields often produce data that can be regarded as resulting from a composition of basic features. Here we present an unsupervised method based on autoencoders for inferring these basic features of data. The main novelty in our approach is that the training is based on the optimization of the `local entropy' rather than the standard loss, resulting in a more robust inference, and enhancing the performance on this type of data considerably. Algorithmically, this is realized by training an interacting system of replicated autoencoders. We apply this method to synthetic and protein sequence data, and show that it is able to infer a hidden representation that correlates well with the underlying generative process, without requiring any prior knowledge.

q-bio.QM↗

arXiv2019

Space-time least-squares isogeometric method and efficient solver for parabolic problems

In this paper, we propose a space-time least-squares isogeometric method to solve parabolic evolution problems, well suited for high-degree smooth splines in the space-time domain. We focus on the linear solver and its computational efficiency: thanks to the proposed formulation and to the tensor-product construction of space-time splines, we can design a preconditioner whose application requires the solution of a Sylvester-like equation, which is performed efficiently by the fast diagonalization method. The preconditioner is robust w.r.t. spline degree and mesh size. The computational time required for its application, for a serial execution, is almost proportional to the number of degrees-of-freedom and independent of the polynomial degree. The proposed approach is also well-suited for parallelization.

math.NA↗

arXiv2019

Multilingual Neural Machine Translation for Zero-Resource Languages

In recent years, Neural Machine Translation (NMT) has been shown to be more effective than phrase-based statistical methods, thus quickly becoming the state of the art in machine translation (MT). However, NMT systems are limited in translating low-resourced languages, due to the significant amount of parallel data that is required to learn useful mappings between languages. In this work, we show how the so-called multilingual NMT can help to tackle the challenges associated with low-resourced language translation. The underlying principle of multilingual NMT is to force the creation of hidden representations of words in a shared semantic space across multiple languages, thus enabling a positive parameter transfer across languages. Along this direction, we present multilingual translation experiments with three languages (English, Italian, Romanian) covering six translation directions, utilizing both recurrent neural networks and transformer (or self-attentive) neural networks. We then focus on the zero-shot translation problem, that is how to leverage multi-lingual data in order to learn translation directions that are not covered by the available training material. To this aim, we introduce our recently proposed iterative self-training method, which incrementally improves a multilingual NMT on a zero-shot direction by just relying on monolingual data. Our results on TED talks data show that multilingual NMT outperforms conventional bilingual NMT, that the transformer NMT outperforms recurrent NMT, and that zero-shot NMT outperforms conventional pivoting methods and even matches the performance of a fully-trained bilingual system.

cs.CL↗

arXiv2019

$Γ$-convergence for high order phase field fracture: continuum and isogeometric formulations

We consider second order phase field functionals, in the continuum setting, and their discretization with isogeometric tensor product B-splines. We prove that these functionals, continuum and discrete, $Γ$-converge to a brittle fracture energy, defined in the space $GSBD^2$. In particular, in the isogeometric setting, since the projection operator is not Lagrangian (i.e., interpolatory) a special construction is needed in order to guarantee that recovery sequences take values in $[0,1]$; convergence holds, as expected, if $h = o (\varepsilon)$, being $h$ the size of the physical mesh and $\varepsilon$ the internal length in the phase field energy.

math.NA↗

arXiv2019

Weak solutions for gradient flows under monotonicity constraints

We consider the gradient flow of a quadratic non-autonomous energy under monotonicity constraint in time and natural regularity assumptions. We provide first a notion of weak solution, inspired by the theory of curves of maximal slope, and then existence (employing time-discrete schemes with different "implementations" of the constraint), uniqueness, power and energy identity, comparison principle and continuous dependence. As a byproduct, we show that the energy identity gives a selection criterion for the (non-unique) evolutions obtained by other notions of solutions. We finally show that, for autonomous energies, the solutions obtained with the monotonicity constraint actually coincide with those obtained with a fixed obstacle, given by the initial datum.

math.AP↗

arXiv2019

Analysis of staggered evolutions for nonlinear energies in phase field fracture

We consider a class of separately convex phase field energies employed in fracture mechanics, featuring non-interpenetration and a general softening behavior. We analyze the time-discrete evolutions generated by a staggered minimization scheme, where fracture irreversibility is modeled by a monotonicity constraint on the phase field variable. After recasting the staggered scheme by means of gradient flows, we characterize the time-continuous limits of the discrete solutions in terms of balanced viscosity evolutions, parametrized by their arc-length with respect to the L2-norm (for the phase field) and the H1-norm (for the displacement field). By a careful study of the energy balance we deduce that time-continuous evolutions may still exhibit an alternate behavior in discontinuity times.

math.AP↗

arXiv2018

Improving Zero-Shot Translation of Low-Resource Languages

Recent work on multilingual neural machine translation reported competitive performance with respect to bilingual models and surprisingly good performance even on (zeroshot) translation directions not observed at training time. We investigate here a zero-shot translation in a particularly lowresource multilingual setting. We propose a simple iterative training procedure that leverages a duality of translations directly generated by the system for the zero-shot directions. The translations produced by the system (sub-optimal since they contain mixed language from the shared vocabulary), are then used together with the original parallel data to feed and iteratively re-train the multilingual network. Over time, this allows the system to learn from its own generated and increasingly better output. Our approach shows to be effective in improving the two zero-shot directions of our multilingual model. In particular, we observed gains of about 9 BLEU points over a baseline multilingual model and up to 2.08 BLEU over a pivoting mechanism using two bilingual models. Further analysis shows that there is also a slight improvement in the non-zero-shot language directions.

cs.CL↗

arXiv2018

Transfer Learning in Multilingual Neural Machine Translation with Dynamic Vocabulary

We propose a method to transfer knowledge across neural machine translation (NMT) models by means of a shared dynamic vocabulary. Our approach allows to extend an initial model for a given language pair to cover new languages by adapting its vocabulary as long as new data become available (i.e., introducing new vocabulary items if they are not included in the initial model). The parameter transfer mechanism is evaluated in two scenarios: i) to adapt a trained single language NMT system to work with a new language pair and ii) to continuously add new language pairs to grow to a multilingual NMT system. In both the scenarios our goal is to improve the translation performance, while minimizing the training convergence time. Preliminary experiments spanning five languages with different training data sizes (i.e., 5k and 50k parallel sentences) show a significant performance gain ranging from +3.85 up to +13.63 BLEU in different language directions. Moreover, when compared with training an NMT model from scratch, our transfer-learning approach allows us to reach higher performance after training up to 4% of the total training steps.

cs.CL↗

arXiv2018

Fine-tuning on Clean Data for End-to-End Speech Translation: FBK @ IWSLT 2018

This paper describes FBK's submission to the end-to-end English-German speech translation task at IWSLT 2018. Our system relies on a state-of-the-art model based on LSTMs and CNNs, where the CNNs are used to reduce the temporal dimension of the audio input, which is in general much higher than machine translation input. Our model was trained only on the audio-to-text parallel data released for the task, and fine-tuned on cleaned subsets of the original training corpus. The addition of weight normalization and label smoothing improved the baseline system by 1.0 BLEU point on our validation set. The final submission also featured checkpoint averaging within a training run and ensemble decoding of models trained during multiple runs. On test data, our best single model obtained a BLEU score of 9.7, while the ensemble obtained a BLEU score of 10.24.

eess.AS↗

arXiv2018

eSCAPE: a Large-scale Synthetic Corpus for Automatic Post-Editing

Training models for the automatic correction of machine-translated text usually relies on data consisting of (source, MT, human post- edit) triplets providing, for each source sentence, examples of translation errors with the corresponding corrections made by a human post-editor. Ideally, a large amount of data of this kind should allow the model to learn reliable correction patterns and effectively apply them at test stage on unseen (source, MT) pairs. In practice, however, their limited availability calls for solutions that also integrate in the training process other sources of knowledge. Along this direction, state-of-the-art results have been recently achieved by systems that, in addition to a limited amount of available training data, exploit artificial corpora that approximate elements of the "gold" training instances with automatic translations. Following this idea, we present eSCAPE, the largest freely-available Synthetic Corpus for Automatic Post-Editing released so far. eSCAPE consists of millions of entries in which the MT element of the training triplets has been obtained by translating the source side of publicly-available parallel corpora, and using the target side as an artificial human post-edit. Translations are obtained both with phrase-based and neural models. For each MT paradigm, eSCAPE contains 7.2 million triplets for English-German and 3.3 millions for English-Italian, resulting in a total of 14,4 and 6,6 million instances respectively. The usefulness of eSCAPE is proved through experiments in a general-domain scenario, the most challenging one for automatic post-editing. For both language directions, the models trained on our artificial data always improve MT quality with statistically significant gains. The current version of eSCAPE can be freely downloaded from: http://hltshare.fbk.eu/QT21/eSCAPE.html.

cs.CL↗