Search Results for author:
Found 49 papers, 24 papers with code
A Natural Diet: Towards Improving Naturalness of Machine Translation Output
In this work we propose a method for training MT systems to achieve a more natural style, i. e. mirroring the style of text originally written in the target language.
High-Resource Methodological Bias in Low-Resource Investigations
The central bottleneck for low-resource NLP is typically regarded to be the quantity of accessible data, overlooking the contribution of data quality.
AudioLM: a Language Modeling Approach to Audio Generation
We introduce AudioLM, a framework for high-quality audio generation with long-term consistency.
Learning strides in convolutional neural networks
In particular, we show that introducing our layer into a ResNet-18 architecture allows keeping consistent high performance on CIFAR10, CIFAR100 and ImageNet even when training starts from poor random stride configurations.
High Quality Rather than High Model Probability: Minimum Bayes Risk Decoding with Neural Metrics
In Neural Machine Translation, it is typically assumed that the sentence with the highest estimated probability should also be the translation with the highest quality as measured by humans.
The Trade-offs of Domain Adaptation for Neural Language Models
This work connects language model adaptation with concepts of machine learning theory.
On the Complementarity of Data Selection and Fine Tuning for Domain Adaptation
Domain adaptation of neural networks commonly relies on three training phases: pretraining, selected data training and then fine tuning.
Auxiliary Task Update Decomposition: The Good, The Bad and The Neutral
In this case, careful consideration is needed to select tasks and model parameterizations such that updates from the auxiliary tasks actually help the primary task.
On Systematic Style Differences between Unsupervised and Supervised MT and an Application for High-Resource Machine Translation
Modern unsupervised machine translation (MT) systems reach reasonable translation quality under clean and controlled data conditions.
DIVE: End-to-end Speech Diarization via Iterative Speaker Embedding
Our neural algorithm presents the diarization task as an iterative process: it repeatedly builds a representation for each speaker before predicting the voice activity of each speaker conditioned on the extracted representations.
Experts, Errors, and Context: A Large-Scale Study of Human Evaluation for Machine Translation
Human evaluation of modern high-quality machine translation systems is a difficult problem, and there is increasing evidence that inadequate evaluation procedures can lead to erroneous conclusions.
Learning from Heterogeneous EEG Signals with Differentiable Channel Reordering
We propose CHARM, a method for training a single neural network across inconsistent input channels.
Contrastive Learning of General-Purpose Audio Representations
We introduce COLA, a self-supervised pre-training approach for learning a general-purpose representation of audio.
Ranked #4 on
Spoken Command Recognition
on Speech Command v2
Human-Paraphrased References Improve Neural Machine Translation
When used in place of original references, the paraphrased versions produce metric scores that correlate better with human judgment.
Translationese as a Language in ``Multilingual'' NMT
Machine translation has an undesirable propensity to produce {``}translationese{''} artifacts, which can lead to higher BLEU scores while being liked less by human raters.
Toward Better Storylines with Sentence-Level Language Models
We propose a sentence-level language model which selects the next sentence in a story from a finite set of fluent alternatives.
BLEU might be Guilty but References are not Innocent
The quality of automatic metrics for machine translation has been increasingly called into question, especially for high-quality systems.
Efficient Content-Based Sparse Attention with Routing Transformers
This work builds upon two lines of research: it combines the modeling flexibility of prior work on content-based sparse attention with the efficiency gains from approaches based on local, temporal sparse attention.
Ranked #5 on
Image Generation
on ImageNet 64x64
(Bits per dim metric)
Wavesplit: End-to-End Speech Separation by Speaker Clustering
Wavesplit infers a set of source representations via clustering, which addresses the fundamental permutation problem of separation.
Ranked #3 on
Speech Separation
on WHAMR!
Translationese as a Language in "Multilingual" NMT
Machine translation has an undesirable propensity to produce "translationese" artifacts, which can lead to higher BLEU scores while being liked less by human raters.
ELI5: Long Form Question Answering
We introduce the first large-scale corpus for long-form question answering, a task requiring elaborate and in-depth answers to open-ended questions.
Tagged Back-Translation
Recent work in Neural Machine Translation (NMT) has shown significant quality gains from noised-beam decoding during back-translation, a method to generate synthetic parallel data.
Unsupervised Hierarchical Story Infilling
Story infilling involves predicting words to go into a missing span from a story.
Unsupervised Paraphrasing without Translation
We compare with MT-based approaches on paraphrase identification, generation, and training augmentation.
fairseq: A Fast, Extensible Toolkit for Sequence Modeling
fairseq is an open-source sequence modeling toolkit that allows researchers and developers to train custom models for translation, summarization, language modeling, and other text generation tasks.
Modeling Human Motion with Quaternion-based Neural Networks
Previous work on predicting or generating 3D human pose sequences regresses either joint rotations or joint positions.
3D human pose estimation in video with temporal convolutions and semi-supervised training
We start with predicted 2D keypoints for unlabeled video, then estimate 3D poses and finally back-project to the input 2D keypoints.
Ranked #11 on
Weakly-supervised 3D Human Pose Estimation
on Human3.6M
(Number of Frames Per View metric)
Monocular 3D Human Pose Estimation
Weakly-supervised 3D Human Pose Estimation
Understanding Back-Translation at Scale
An effective method to improve neural machine translation with monolingual data is to augment the parallel training corpus with back-translations of target language sentences.
Ranked #2 on
Machine Translation
on WMT2014 English-German
(using extra training data)
QuickEdit: Editing Text \& Translations by Crossing Words Out
We also evaluate our model for paraphrasing through a user study.
Scaling Neural Machine Translation
Sequence to sequence learning models still require several days to reach state of the art performance on large benchmark datasets using a single machine.
Ranked #12 on
Machine Translation
on WMT2014 English-French
QuaterNet: A Quaternion-based Recurrent Model for Human Motion
Deep learning for predicting or generating 3D human pose sequences is an active research area.
Analyzing Uncertainty in Neural Machine Translation
We propose tools and metrics to assess how uncertainty in the data is captured by the model distribution and how it affects search strategies that generate translations.
Classical Structured Prediction Losses for Sequence to Sequence Learning
There has been much recent work on training neural attention models at the sequence-level using either reinforcement learning-style methods or by optimizing the beam.
Ranked #4 on
Machine Translation
on IWSLT2015 German-English
Controllable Abstractive Summarization
Current models for document summarization disregard user preferences such as the desired length, style, the entities that the user might be interested in, or how much of the document the user has already read.
QuickEdit: Editing Text & Translations by Crossing Words Out
We also evaluate our model for paraphrasing through a user study.
Convolutional Sequence to Sequence Learning
The prevalent approach to sequence to sequence learning maps an input sequence to a variable length output sequence via recurrent neural networks.
Language Modeling with Gated Convolutional Networks
The pre-dominant approach to language modeling to date is based on recurrent neural networks.
Ranked #18 on
Language Modelling
on One Billion Word
A Convolutional Encoder Model for Neural Machine Translation
The prevalent approach to neural machine translation relies on bi-directional LSTMs to encode the source sentence.
Ranked #7 on
Machine Translation
on IWSLT2015 German-English
Iterative Refinement for Machine Translation
Existing machine translation decoding algorithms generate translations in a strictly monotonic fashion and never revisit previous decisions.
Vocabulary Selection Strategies for Neural Machine Translation
Classical translation models constrain the space of possible outputs by selecting a subset of translation rules based on the input sentence.
Efficient softmax approximation for GPUs
We propose an approximate strategy to efficiently train neural network based language models over very large vocabularies.
Interactive Semantic Featuring for Text Classification
In text classification, dictionaries can be used to define human-comprehensible features.
Neural Text Generation from Structured Data with Application to the Biography Domain
This paper introduces a neural model for concept-to-text generation that scales to large, rich domains.
Ranked #4 on
Table-to-Text Generation
on WikiBio
Strategies for Training Large Vocabulary Neural Language Models
Training neural network language models over large vocabularies is still computationally very costly compared to count-based models such as Kneser-Ney.
Predicting distributions with Linearizing Belief Networks
Contrary to a classical neural network, a belief network can predict more than the expected value of the output $Y$ given the input $X$.
ICE: Enabling Non-Experts to Build Models Interactively for Large-Scale Lopsided Problems
Based on the machine's output, the teacher can revise the definition of the task or make it more precise.
Label Embedding Trees for Large Multi-Class Tasks
Multi-class classification becomes challenging at test time when the number of classes is very large and testing against every possible class can become computationally infeasible.
Feature Set Embedding for Incomplete Data
Our proposal maps (feature, value) pairs into an embedding space and then non-linearly combines the set of embedded vectors.
Polynomial Semantic Indexing
We present a class of nonlinear (polynomial) models that are discriminatively trained to directly map from the word content in a query-document or document-document pair to a ranking score.
