arXiv subjects

Hao Xiong

Publications and source records attributed to Hao Xiong.

46 records · Page 3Linked to original sources

arXiv2019

Robust Neural Machine Translation with Joint Textual and Phonetic Embedding

Neural machine translation (NMT) is notoriously sensitive to noises, but noises are almost inevitable in practice. One special kind of noise is the homophone noise, where words are replaced by other words with similar pronunciations. We propose to improve the robustness of NMT to homophone noises by 1) jointly embedding both textual and phonetic information of source sentences, and 2) augmenting the training dataset with homophone noises. Interestingly, to achieve better translation quality and more robustness, we found that most (though not all) weights should be put on the phonetic rather than textual information. Experiments show that our method not only significantly improves the robustness of NMT to homophone noises, but also surprisingly improves the translation quality on some clean test sets.

cs.CL↗

arXiv2019

End-to-End Speech Translation with Knowledge Distillation

End-to-end speech translation (ST), which directly translates from source language speech into target language text, has attracted intensive attentions in recent years. Compared to conventional pipeline systems, end-to-end ST models have advantages of lower latency, smaller model size and less error propagation. However, the combination of speech recognition and text translation in one model is more difficult than each of these two tasks. In this paper, we propose a knowledge distillation approach to improve ST model by transferring the knowledge from text translation model. Specifically, we first train a text translation model, regarded as a teacher model, and then ST model is trained to learn output probabilities from teacher model through knowledge distillation. Experiments on English- French Augmented LibriSpeech and English-Chinese TED corpus show that end-to-end ST is possible to implement on both similar and dissimilar language pairs. In addition, with the instruction of teacher model, end-to-end ST model can gain significant improvements by over 3.5 BLEU points.

cs.CL↗

arXiv2018

Phase-mediated magnon chaos-order transition in cavity optomagnonics

Magnon as a quantized spin wave has attracted extensive attentions in various fields of physics, such as magnon spintronics, microwave photonics, and cavity quantum electrodynamics. Here, we explore theoretically magnon chaos-order transition in cavity optomagnonics, which is still remains largely unexplored in this emerging field. We find that the evolution of the magnon experiences the transition from order to period-doubling bifurcation and finally enter chaos by adjusting the microwave driving power. Different from the normal chaos, the magnon chaos-order transition proposed here is phase-mediated. Beyond their fundamental scientific significance, our results will contribute to the comprehending of nonlinear phenomena and chaos in optomagnonical system, and may find applications in chaos-based secure communication.

cond-mat.mes-hall↗

arXiv2018

Modeling Coherence for Discourse Neural Machine Translation

Discourse coherence plays an important role in the translation of one text. However, the previous reported models most focus on improving performance over individual sentence while ignoring cross-sentence links and dependencies, which affects the coherence of the text. In this paper, we propose to use discourse context and reward to refine the translation quality from the discourse perspective. In particular, we generate the translation of individual sentences at first. Next, we deliberate the preliminary produced translations, and train the model to learn the policy that produces discourse coherent text by a reward teacher. Practical results on multiple discourse test datasets indicate that our model significantly improves the translation quality over the state-of-the-art baseline system by +1.23 BLEU score. Moreover, our model generates more discourse coherent text and obtains +2.2 BLEU improvements when evaluated by discourse metrics.

cs.CL↗

arXiv2018

Magnon-induced high-order sideband generation

Magnon Kerr nonlinearity plays crucial roles in the study of cavity optomagnonics system and may bring many novel physical phenomena and important applications. In this work, we report the investigation of high- order sideband generation induced by magnon Kerr nonlinearity in a cavity-magnon system, which is critically lacking in this emerging research field. We uncover that the driving field plays a significant role in controlling the generation and amplification of the higher-order sidebands and the sideband spacing can be adjusted by regulating the beat frequency between the pump laser and the probe laser, which is extremely eventful for the spacing modulation of the sideband spectrum. Based on the recent experimental progress, our results deepen our cognition into optomagnonics nonlinearity and may find interesting applications in optical frequency metrology and optical communications.

physics.optics↗

arXiv2018

Optomechanical Akhmediev Breather

Akhmediev breather has attracted considerable attention over the past decades in various fields such as hydrodynamics, plasma physics, and nonlinear optics because its wide applications and its importance in understanding nonlinear coherent phenomena. Here we introduce optomechanical arrays as a novel and integrated platform to observe both photonic and phononic Akhmediev breather, with remarkable functionality for readout and control complementing existing research on hydrodynamic and plasma systems. Our results pave a path toward reconfigurable nano- photonic and phononic networks, and may enable a new class of devices in signal processing and on-chip nonlinear dynamical systems.

physics.optics↗

arXiv2017

Generation and amplification of high-order sideband induced by two-level atoms in a hybrid optomechanical system

It is quite important to enhance and control the optomechanically induced high-order sideband generation to achieve low-power optical comb and high-sensitivity sensing with an integratable structure. Here we present and analyze a proposal for enhancement and manipulation of optical nonlinearity and high-order sideband generation in a hybrid atom-cavity optomechanical system that is coherently driven by a bichromatic input field consisting of a control field and a probe field and works beyond the perturbative regime. Our numerical analysis with experimentally achievable parameters confirms that robust high-order sideband generation and typical spectral structures with non-perturbative features can be created even under weak driven fields. The dependence of the high-order sideband generation on the atomic parameters are also discussed in detail, including the decay rate of the atoms and the coupling parameter between the atoms and the cavity field. We show that the cutoff order as well as the amplitude of the higher order sidebands can be well tuned by the atomic coupling strength and the atomic decay rate. The proposed mechanism of enhancing optical nonlinearity is quite general and can be adopted to optomechanical systems with different types of cavity.

physics.optics↗

arXiv2017

Multi-channel Encoder for Neural Machine Translation

Attention-based Encoder-Decoder has the effective architecture for neural machine translation (NMT), which typically relies on recurrent neural networks (RNN) to build the blocks that will be lately called by attentive reader during the decoding process. This design of encoder yields relatively uniform composition on source sentence, despite the gating mechanism employed in encoding RNN. On the other hand, we often hope the decoder to take pieces of source sentence at varying levels suiting its own linguistic structure: for example, we may want to take the entity name in its raw form while taking an idiom as a perfectly composed unit. Motivated by this demand, we propose Multi-channel Encoder (MCE), which enhances encoding components with different levels of composition. More specifically, in addition to the hidden state of encoding RNN, MCE takes 1) the original word embedding for raw encoding with no composition, and 2) a particular design of external memory in Neural Turing Machine (NTM) for more complex composition, while all three encoding strategies are properly blended during decoding. Empirical study on Chinese-English translation shows that our model can improve by 6.52 BLEU points upon a strong open source NMT system: DL4MT1. On the WMT14 English- French task, our single shallow system achieves BLEU=38.8, comparable with the state-of-the-art deep models.

cs.CL↗

arXiv2017

Highly sensitive optical sensor for precision measurement of electrical charges based on optomechanically induced difference-sideband generation

Difference-sideband generation in an optomechanical system coupled to a charged object is investigated beyond the conventional linearized description of optomechanical interactions. An exponential decay law for difference-sideband generation in the presence of electric interaction is identified which exhibits more sensitivity to electrical charges than the conventional linearized effects. Using exact the same parameters with previous work based on the linearized dynamics of the optomechanical interactions, we show that optomechanically induced difference-sideband generation may enable an all-optical sensor for precision measurement of electrical charges with higher precision and lower power. The proposed mechanism is especially suited for on-chip optomechanical devices, where nonlinear optomechanical interaction in the weak coupling regime is within current experimental reach.

physics.ins-det↗

arXiv2016

Radiation pressure induced difference-sideband generation beyond linearized description

We investigate radiation-pressure induced generation of the frequency components at the difference-sideband in an optomechanical system, which beyond the conventional linearized description of optomechanical interactions between cavity fields and the mechanical oscillation. We analytically calculate amplitudes of these signals, and identify a simple square-root law for both the upper and lower difference-sideband generation which can describe the dependence of the intensities of these signals on the pump power. Further calculation shows that difference-sideband generation can be greatly enhanced via achieving the matching conditions. The effect of difference-sideband generation, which may have potential application for manipulation of light, is especially suited for on-chip optomechanical devices, where nonlinear optomechanical interaction in the weak coupling regime is within current experimental reach.

physics.optics↗