arXiv subjects

Biao Zhang

Publications and source records attributed to Biao Zhang.

At least 91 records · Page 5Linked to original sources

arXiv2020

Mid-infrared single photon detector with superconductor Mo$_{80}$Si$_{20}$ nanowire

A mid-infrared single photon detector (MIR-SNSPD) was reported based on 30 nm-wide superconductor molybdenum silicide nanowires in this work. Saturated quantum efficiencies (QEs) were achieved at the wavelength ranging from 1.55 to 5.07 micrometer in experiments. At the same time, the intrinsic dark count rate (DCR) was below 100 cps. Thus, this device produced a noise equivalent power (NEP) of 4.5 * 10-19 W/sqrt(Hz). The results provide the foundation of developing 10 micrometer-SNSPD for the applications of infrared astronomy observation.

physics.app-ph↗

arXiv2020

Fast Interleaved Bidirectional Sequence Generation

Independence assumptions during sequence generation can speed up inference, but parallel generation of highly inter-dependent tokens comes at a cost in quality. Instead of assuming independence between neighbouring tokens (semi-autoregressive decoding, SA), we take inspiration from bidirectional sequence generation and introduce a decoder that generates target words from the left-to-right and right-to-left directions simultaneously. We show that we can easily convert a standard architecture for unidirectional decoding into a bidirectional decoder by simply interleaving the two directions and adapting the word positions and self-attention masks. Our interleaved bidirectional decoder (IBDecoder) retains the model simplicity and training efficiency of the standard Transformer, and on five machine translation tasks and two document summarization tasks, achieves a decoding speedup of ~2X compared to autoregressive decoding with comparable quality. Notably, it outperforms left-to-right SA because the independence assumptions in IBDecoder are more felicitous. To achieve even higher speedups, we explore hybrid models where we either simultaneously predict multiple neighbouring tokens per direction, or perform multi-directional decoding by partitioning the target sequence. These methods achieve speedups to 4X-11X across different tasks at the cost of <1 BLEU or <0.5 ROUGE (on average). Source code is released at https://github.com/bzhangGo/zero.

cs.CL↗

arXiv2020

Adaptive Feature Selection for End-to-End Speech Translation

Information in speech signals is not evenly distributed, making it an additional challenge for end-to-end (E2E) speech translation (ST) to learn to focus on informative features. In this paper, we propose adaptive feature selection (AFS) for encoder-decoder based E2E ST. We first pre-train an ASR encoder and apply AFS to dynamically estimate the importance of each encoded speech feature to SR. A ST encoder, stacked on top of the ASR encoder, then receives the filtered features from the (frozen) ASR encoder. We take L0DROP (Zhang et al., 2020) as the backbone for AFS, and adapt it to sparsify speech features with respect to both temporal and feature dimensions. Results on LibriSpeech En-Fr and MuST-C benchmarks show that AFS facilitates learning of ST by pruning out ~84% temporal features, yielding an average translation gain of ~1.3-1.6 BLEU and a decoding speedup of ~1.4x. In particular, AFS reduces the performance gap compared to the cascade baseline, and outperforms it on LibriSpeech En-Fr with a BLEU score of 18.56 (without data augmentation)

cs.CL↗

arXiv2020

A Study in Improving BLEU Reference Coverage with Diverse Automatic Paraphrasing

We investigate a long-perceived shortcoming in the typical use of BLEU: its reliance on a single reference. Using modern neural paraphrasing techniques, we study whether automatically generating additional diverse references can provide better coverage of the space of valid translations and thereby improve its correlation with human judgments. Our experiments on the into-English language directions of the WMT19 metrics task (at both the system and sentence level) show that using paraphrased references does generally improve BLEU, and when it does, the more diverse the better. However, we also show that better results could be achieved if those paraphrases were to specifically target the parts of the space most relevant to the MT outputs being evaluated. Moreover, the gains remain slight even when human paraphrases are used, suggesting inherent limitations to BLEU's capacity to correctly exploit multiple references. Surprisingly, we also find that adequacy appears to be less important, as shown by the high results of a strong sampling approach, which even beats human paraphrases when used with sentence-level BLEU.

cs.CL↗

arXiv2020

Structural Multi-Colour Invisible Inks with Submicron 4D Printing of Shape Memory Polymers

Four-dimensional (4D) printing of shape memory polymer (SMP) imparts time responsive properties to 3D structures. Here, we explore 4D printing of a SMP in the submicron length scale, extending its applications to nanophononics. We report a new SMP photoresist based on Vero Clear achieving print features at a resolution of ~300 nm half pitch using two-photon polymerization lithography (TPL). Prints consisting of grids with size-tunable multi-colours enabled the study of shape memory effects to achieve large visual shifts through nanoscale structure deformation. As the nanostructures are flattened, the colours and printed information become invisible. Remarkably, the shape memory effect recovers the original surface morphology of the nanostructures along with its structural colour within seconds of heating above its glass transition temperature. The high-resolution printing and excellent reversibility in both microtopography and optical properties promises a platform for temperature-sensitive labels, information hiding for anti-counterfeiting, and tunable photonic devices.

physics.app-ph↗

arXiv2020

On Sparsifying Encoder Outputs in Sequence-to-Sequence Models

Sequence-to-sequence models usually transfer all encoder outputs to the decoder for generation. In this work, by contrast, we hypothesize that these encoder outputs can be compressed to shorten the sequence delivered for decoding. We take Transformer as the testbed and introduce a layer of stochastic gates in-between the encoder and the decoder. The gates are regularized using the expected value of the sparsity-inducing L0penalty, resulting in completely masking-out a subset of encoder outputs. In other words, via joint training, the L0DROP layer forces Transformer to route information through a subset of its encoder states. We investigate the effects of this sparsification on two machine translation and two summarization tasks. Experiments show that, depending on the task, around 40-70% of source encodings can be pruned without significantly compromising quality. The decrease of the output length endows L0DROP with the potential of improving decoding efficiency, where it yields a speedup of up to 1.65x on document summarization tasks against the standard Transformer. We analyze the L0DROP behaviour and observe that it exhibits systematic preferences for pruning certain word types, e.g., function words and punctuation get pruned most. Inspired by these observations, we explore the feasibility of specifying rule-based patterns that mask out encoder outputs based on information such as part-of-speech tags, word frequency and word position.

cs.CL↗

arXiv2020

Improving Massively Multilingual Neural Machine Translation and Zero-Shot Translation

Massively multilingual models for neural machine translation (NMT) are theoretically attractive, but often underperform bilingual models and deliver poor zero-shot translations. In this paper, we explore ways to improve them. We argue that multilingual NMT requires stronger modeling capacity to support language pairs with varying typological characteristics, and overcome this bottleneck via language-specific components and deepening NMT architectures. We identify the off-target translation issue (i.e. translating into a wrong target language) as the major source of the inferior zero-shot performance, and propose random online backtranslation to enforce the translation of unseen training language pairs. Experiments on OPUS-100 (a novel multilingual dataset with 100 languages) show that our approach substantially narrows the performance gap with bilingual models in both one-to-many and many-to-many settings, and improves zero-shot performance by ~10 BLEU, approaching conventional pivot-based methods.

cs.CL↗

arXiv2019

A GRU-Gated Attention Model for Neural Machine Translation

Neural machine translation (NMT) heavily relies on an attention network to produce a context vector for each target word prediction. In practice, we find that context vectors for different target words are quite similar to one another and therefore are insufficient in discriminatively predicting target words. The reason for this might be that context vectors produced by the vanilla attention network are just a weighted sum of source representations that are invariant to decoder states. In this paper, we propose a novel GRU-gated attention model (GAtt) for NMT which enhances the degree of discrimination of context vectors by enabling source representations to be sensitive to the partial translation generated by the decoder. GAtt uses a gated recurrent unit (GRU) to combine two types of information: treating a source annotation vector originally produced by the bidirectional encoder as the history state while the corresponding previous decoder state as the input to the GRU. The GRU-combined information forms a new source annotation vector. In this way, we can obtain translation-sensitive source representations which are then feed into the attention network to generate discriminative context vectors. We further propose a variant that regards a source annotation vector as the current input while the previous decoder state as the history. Experiments on NIST Chinese-English translation tasks show that both GAtt-based models achieve significant improvements over the vanilla attentionbased NMT. Further analyses on attention weights and context vectors demonstrate the effectiveness of GAtt in improving the discrimination power of representations and handling the challenging issue of over-translation.

cs.CL↗

arXiv2019

Root Mean Square Layer Normalization

Layer normalization (LayerNorm) has been successfully applied to various deep neural networks to help stabilize training and boost model convergence because of its capability in handling re-centering and re-scaling of both inputs and weight matrix. However, the computational overhead introduced by LayerNorm makes these improvements expensive and significantly slows the underlying network, e.g. RNN in particular. In this paper, we hypothesize that re-centering invariance in LayerNorm is dispensable and propose root mean square layer normalization, or RMSNorm. RMSNorm regularizes the summed inputs to a neuron in one layer according to root mean square (RMS), giving the model re-scaling invariance property and implicit learning rate adaptation ability. RMSNorm is computationally simpler and thus more efficient than LayerNorm. We also present partial RMSNorm, or pRMSNorm where the RMS is estimated from p% of the summed inputs without breaking the above properties. Extensive experiments on several tasks using diverse network architectures show that RMSNorm achieves comparable performance against LayerNorm but reduces the running time by 7%~64% on different models. Source code is available at https://github.com/bzhangGo/rmsnorm.

cs.LG↗

arXiv2019

Improving Deep Transformer with Depth-Scaled Initialization and Merged Attention

The general trend in NLP is towards increasing model capacity and performance via deeper neural networks. However, simply stacking more layers of the popular Transformer architecture for machine translation results in poor convergence and high computational overhead. Our empirical analysis suggests that convergence is poor due to gradient vanishing caused by the interaction between residual connections and layer normalization. We propose depth-scaled initialization (DS-Init), which decreases parameter variance at the initialization stage, and reduces output variance of residual connections so as to ease gradient back-propagation through normalization layers. To address computational cost, we propose a merged attention sublayer (MAtt) which combines a simplified averagebased self-attention sublayer and the encoderdecoder attention sublayer on the decoder side. Results on WMT and IWSLT translation tasks with five translation directions show that deep Transformers with DS-Init and MAtt can substantially outperform their base counterpart in terms of BLEU (+1.1 BLEU on average for 12-layer models), while matching the decoding speed of the baseline model thanks to the efficiency improvements of MAtt.

cs.CL↗

arXiv2019

A Lightweight Recurrent Network for Sequence Modeling

Recurrent networks have achieved great success on various sequential tasks with the assistance of complex recurrent units, but suffer from severe computational inefficiency due to weak parallelization. One direction to alleviate this issue is to shift heavy computations outside the recurrence. In this paper, we propose a lightweight recurrent network, or LRN. LRN uses input and forget gates to handle long-range dependencies as well as gradient vanishing and explosion, with all parameter related calculations factored outside the recurrence. The recurrence in LRN only manipulates the weight assigned to each token, tightly connecting LRN with self-attention networks. We apply LRN as a drop-in replacement of existing recurrent units in several neural sequential models. Extensive experiments on six NLP tasks show that LRN yields the best running efficiency with little or no loss in model performance.

cs.CL↗

arXiv2019

Revisiting Low-Resource Neural Machine Translation: A Case Study

It has been shown that the performance of neural machine translation (NMT) drops starkly in low-resource conditions, underperforming phrase-based statistical machine translation (PBSMT) and requiring large amounts of auxiliary data to achieve competitive results. In this paper, we re-assess the validity of these results, arguing that they are the result of lack of system adaptation to low-resource settings. We discuss some pitfalls to be aware of when training low-resource NMT systems, and recent techniques that have shown to be especially helpful in low-resource settings, resulting in a set of best practices for low-resource NMT. In our experiments on German--English with different amounts of IWSLT14 training data, we show that, without the use of any auxiliary monolingual or multilingual data, an optimized NMT system can outperform PBSMT with far less data than previously claimed. We also apply these techniques to a low-resource Korean-English dataset, surpassing previously reported results by 4 BLEU.

cs.CL↗

arXiv2018

Cseq2seq: Cyclic Sequence-to-Sequence Learning

The vanilla sequence-to-sequence learning (seq2seq) reads and encodes a source sequence into a fixed-length vector only once, suffering from its insufficiency in modeling structural correspondence between the source and target sequence. Instead of handling this insufficiency with a linearly weighted attention mechanism, in this paper, we propose to use a recurrent neural network (RNN) as an alternative (Cseq2seq-I). During decoding, Cseq2seq-I cyclically feeds the previous decoding state back to the encoder as the initial state of the RNN, and reencodes source representations to produce context vectors. We surprisingly find that the introduced RNN succeeds in dynamically detecting translationrelated source tokens according to the partial target sequence. Based on this finding, we further hypothesize that the partial target sequence can act as a feedback to improve the understanding of the source sequence. To test this hypothesis, we propose cyclic sequence-to-sequence learning (Cseq2seq-II) which differs from the seq2seq only in the reintroduction of previous decoding state into the same encoder. We further perform parameter sharing on Cseq2seq-II to reduce parameter redundancy and enhance regularization. In particular, we share the weights of the encoder and decoder, and two targetside word embeddings, making Cseq2seq-II equivalent to a single conditional RNN model, with 31% parameters pruned but even better performance. Cseq2seq-II not only preserves the simplicity of seq2seq but also yields comparable and promising results on machine translation tasks. Experiments on Chinese- English and English-German translation show that Cseq2seq achieves significant and consistent improvements over seq2seq and is as competitive as the attention-based seq2seq model.

cs.CL↗

arXiv2018

A 16-pixel NbN nanowire single photon detector coupled with 300 micrometer fiber

Niobium Nitride (NbN) nanowire is the most popular detection material of superconducting nanowire single photon detectors (SNSPDs) for high repetition rate and high efficiency. However, it has been assumed to be difficult for fabricating SNSPDs with arrays over large area, which are critical components of quantum imaging, linear optical quantum computing, satellite laser ranging, and high-speed quantum key distribution. This paper reported a 4*4 pixel NbN SNSPDs array with an equivalent receiving aperture of 300 micrometer associated with beam compression technology, which is the first NbN SNSPD coupled by 300 micrometer fiber according to our knowledge. The designed pixel was compact with a filling factor of 98.5%, resulting in a high quantum efficiency of 94.5%, a system efficiency of 46% for photons coupled from 300 micrometer fiber without optimizing polarization, and a system time resolution of 92 ps. This work demonstrates the feasibility of a large SNSPDs array, and paves the way for developing efficient photon camera with NbN nanowires.

physics.app-ph↗

arXiv2018

Planar penta-transition metal phosphide and arsenide as narrow-bandgap semiconductors from first principle calculations

Searching for materials with single atom-thin as well as planar structure, like graphene and borophene, is one of the most attractive themes in two dimensional materials. Herein, using density functional theory calculations, we have proposed a series of single layer planar penta-transition metal phosphide and arsenide, i.e. TM$\mathrm{_2}$X$\mathrm{_4}$ (TM= Ni, Pd and Pt; X=P, As). According to the calculated phonon dispersion relation and elastic constants, as well as ab initio molecular dynamics simulation results, monolayers of planar penta-TM$\mathrm{_2}$X$\mathrm{_4}$ are dynamically, mechanically, and thermally stable. In addition, the band structures calculated with the screened HSE06 hybrid functional including spin-orbit coupling show that these monolayers are direct-gap semiconductors with sizeable band gaps ranging from 0.14 eV to 0.69 eV. Besides, the optical properties in these monolayers are further investigated, where strong in-plane optical absorption with wide spectral range has been revealed. Our results indicate that planar penta-TM$\mathrm{_2}$X$\mathrm{_4}$ monolayers are interesting narrow gap semiconductors with excellent optical properties, and may find potential applications in photoelectronics.

cond-mat.mtrl-sci↗

arXiv2018

Simplifying Neural Machine Translation with Addition-Subtraction Twin-Gated Recurrent Networks

In this paper, we propose an additionsubtraction twin-gated recurrent network (ATR) to simplify neural machine translation. The recurrent units of ATR are heavily simplified to have the smallest number of weight matrices among units of all existing gated RNNs. With the simple addition and subtraction operation, we introduce a twin-gated mechanism to build input and forget gates which are highly correlated. Despite this simplification, the essential non-linearities and capability of modeling long-distance dependencies are preserved. Additionally, the proposed ATR is more transparent than LSTM/GRU due to the simplification. Forward self-attention can be easily established in ATR, which makes the proposed network interpretable. Experiments on WMT14 translation tasks demonstrate that ATR-based neural machine translation can yield competitive performance on English- German and English-French language pairs in terms of both translation quality and speed. Further experiments on NIST Chinese-English translation, natural language inference and Chinese word segmentation verify the generality and applicability of ATR on different natural language processing tasks.

cs.CL↗

arXiv2018

Otem&Utem: Over- and Under-Translation Evaluation Metric for NMT

Although neural machine translation(NMT) yields promising translation performance, it unfortunately suffers from over- and under-translation is- sues [Tu et al., 2016], of which studies have become research hotspots in NMT. At present, these studies mainly apply the dominant automatic evaluation metrics, such as BLEU, to evaluate the overall translation quality with respect to both adequacy and uency. However, they are unable to accurately measure the ability of NMT systems in dealing with the above-mentioned issues. In this paper, we propose two quantitative metrics, the Otem and Utem, to automatically evaluate the system perfor- mance in terms of over- and under-translation respectively. Both metrics are based on the proportion of mismatched n-grams between gold ref- erence and system translation. We evaluate both metrics by comparing their scores with human evaluations, where the values of Pearson Cor- relation Coefficient reveal their strong correlation. Moreover, in-depth analyses on various translation systems indicate some inconsistency be- tween BLEU and our proposed metrics, highlighting the necessity and significance of our metrics.

cs.CL↗

arXiv2018

Accelerating Neural Transformer via an Average Attention Network

With parallelizable attention networks, the neural Transformer is very fast to train. However, due to the auto-regressive architecture and self-attention in the decoder, the decoding procedure becomes slow. To alleviate this issue, we propose an average attention network as an alternative to the self-attention network in the decoder of the neural Transformer. The average attention network consists of two layers, with an average layer that models dependencies on previous positions and a gating layer that is stacked over the average layer to enhance the expressiveness of the proposed attention network. We apply this network on the decoder part of the neural Transformer to replace the original target-side self-attention model. With masking tricks and dynamic programming, our model enables the neural Transformer to decode sentences over four times faster than its original version with almost no loss in training time and translation performance. We conduct a series of experiments on WMT17 translation tasks, where on 6 different language pairs, we obtain robust and consistent speed-ups in decoding.

cs.CL↗