arXiv subjects

Xinyi Wang

Publications and source records attributed to Xinyi Wang.

At least 235 records · Page 13Linked to original sources

arXiv2021

Identifying latent shared mobility preference segments in low-income communities: ride-hailing, fixed-route bus, and mobility-on-demand transit

Concepts of Mobility-on-Demand (MOD) and Mobility as a Service (MaaS), which feature the integration of various shared-use mobility options, have gained widespread popularity in recent years. While these concepts promise great benefits to travelers, their heavy reliance on technology raises equity concerns as socially disadvantaged population groups can be left out in an era of on-demand mobility. This paper investigates the potential uptake of MOD transit services (integrated fixed-route and on-demand services) among travelers living in low-income communities. Specially, we analyze people's latent attitude towards three shared-use mobility services, including ride-hailing services, fixed-route transit, and MOD transit. We conduct a latent class cluster analysis of 825 survey respondents sampled from low-income neighborhoods in Detroit and Ypsilanti, Michigan. We identified three latent segments: shared-mode enthusiast, shared-mode opponent, and fixed-route transit loyalist. People from the shared-mode enthusiast segment often use ride-hailing services and live in areas with poor transit access, and they are likely to be the early adopters of MOD transit services. The shared-mode opponent segment mainly includes vehicle owners who lack interests in shared mobility options. The fixed-route transit loyalist segment includes a considerable share of low-income individuals who face technological barriers to use the MOD transit. We also find that males, college graduates, car owners, people with a mobile data plan, and people living in poor-transit-access areas have a higher level of preferences for MOD transit services. We conclude with policy recommendations for developing more accessible and equitable MOD transit services.

physics.soc-ph↗

arXiv2021

Applications of Artificial Intelligence to aid detection of dementia: a narrative review on current capabilities and future directions

With populations ageing, the number of people with dementia worldwide is expected to triple to 152 million by 2050. Seventy percent of cases are due to Alzheimer's disease (AD) pathology and there is a 10-20 year 'pre-clinical' period before significant cognitive decline occurs. We urgently need, cost effective, objective methods to detect AD, and other dementias, at an early stage. Risk factor modification could prevent 40% of cases and drug trials would have greater chances of success if participants are recruited at an earlier stage. Currently, detection of dementia is largely by pen and paper cognitive tests but these are time consuming and insensitive to pre-clinical phases. Specialist brain scans and body fluid biomarkers can detect the earliest stages of dementia but are too invasive or expensive for widespread use. With the advancement of technology, Artificial Intelligence (AI) shows promising results in assisting with detection of early-stage dementia. Existing AI-aided methods and potential future research directions are reviewed and discussed.

cs.AI↗

arXiv2021

Multi-view Subword Regularization

Multilingual pretrained representations generally rely on subword segmentation algorithms to create a shared multilingual vocabulary. However, standard heuristic algorithms often lead to sub-optimal segmentation, especially for languages with limited amounts of data. In this paper, we take two major steps towards alleviating this problem. First, we demonstrate empirically that applying existing subword regularization methods(Kudo, 2018; Provilkov et al., 2020) during fine-tuning of pre-trained multilingual representations improves the effectiveness of cross-lingual transfer. Second, to take full advantage of different possible input segmentations, we propose Multi-view Subword Regularization (MVR), a method that enforces the consistency between predictions of using inputs tokenized by the standard and probabilistic segmentations. Results on the XTREME multilingual benchmark(Hu et al., 2020) show that MVR brings consistent improvements of up to 2.5 points over using standard segmentation algorithms.

cs.CL↗

arXiv2021

Adaptive Subband Compression for Streaming of Continuous Point-on-Wave and PMU Data

A data compression system capable of providing real-time streaming of high-resolution continuous point-on-wave (CPOW) and phasor measurement unit (PMU) measurements is proposed. Referred to as adaptive subband compression (ASBC), the proposed technique partitions the signal space into subbands and adaptively compresses subband signals based on each subband's active bandwidth. The proposed technique conforms to existing industry phasor measurement standards, making it suitable for streaming high-resolution CPOW and PMU data either in continuous or burst on-demand/event-triggered modes. Experiments on synthetic and real data show that ASBC reduces the CPOW sampling rates by several orders of magnitude for real-time streaming while maintaining the precision required by industry standards.

eess.SP↗

arXiv2021

Dynamic Heterogeneity, Cooperative Motion, and Johari-Goldstein $β$-Relaxation in a Metallic Glass-Forming Material Exhibiting a Fragile to Strong Transition

We investigate the Johari-Goldstein (JG) $β$-relaxation process in a model metallic glass-forming (GF) material (Al90Sm10), previously studied extensively by both frequency-dependent mechanical measurements and simulation studies devoted to equilibrium properties, by molecular dynamics simulations based on validated and optimized interatomic potentials with the primary aim of better understanding the nature of this universal relaxation process from a dynamic heterogeneity (DH) perspective. The present relatively low temperature and long-time simulations reveal a direct correspondence between the JG $β$-relaxation time $τ_{JG}$ and the lifetime of the mobile particle clusters $τ_M$, defined as in previous DH studies, a relationship dual to the corresponding previously observed relationship between the $α$-relaxation time $τ_α$ and the lifetime of immobile particle clusters $τ_{IM}$. Moreover, we find that the average diffusion coefficient D nearly coincides with D$_{Al}$, of the smaller atomic species (Al), and that the hopping time associated with D coincides with $τ_{JG}$ to within numerical uncertainty, both trends being in accord with experimental studies. This indicates that the JG $α$-relaxation is dominated by the smaller atomic species and the observation of a direct relation between this relaxation process and rate of molecular diffusion in GF materials at low temperatures where the JG $β$-relaxation becomes the prevalent mode of structural relaxation. As an unanticipated aspect of our study, we find that Al90Sm10 exhibits fragile-to-strong (FS) glass-formation, as found in many other metallic GF liquids, but this fact does not greatly alter the geometrical nature of DH in this material and the relation of DH to dynamical properties.

cond-mat.mtrl-sci↗

arXiv2021

Gradient-guided Loss Masking for Neural Machine Translation

To mitigate the negative effect of low quality training data on the performance of neural machine translation models, most existing strategies focus on filtering out harmful data before training starts. In this paper, we explore strategies that dynamically optimize data usage during the training process using the model's gradients on a small set of clean data. At each training step, our algorithm calculates the gradient alignment between the training data and the clean data to mask out data with negative alignment. Our method has a natural intuition: good training data should update the model parameters in a similar direction as the clean data. Experiments on three WMT language pairs show that our method brings significant improvement over strong baselines, and the improvements are generalizable across test data from different domains.

cs.CL↗

arXiv2021

Meta Back-translation

Back-translation is an effective strategy to improve the performance of Neural Machine Translation~(NMT) by generating pseudo-parallel data. However, several recent works have found that better translation quality of the pseudo-parallel data does not necessarily lead to better final translation models, while lower-quality but more diverse data often yields stronger results. In this paper, we propose a novel method to generate pseudo-parallel data from a pre-trained back-translation model. Our method is a meta-learning algorithm which adapts a pre-trained back-translation model so that the pseudo-parallel data it generates would train a forward-translation model to do well on a validation set. In our evaluations in both the standard datasets WMT En-De'14 and WMT En-Fr'14, as well as a multilingual translation setting, our method leads to significant improvements over strong baselines. Our code will be made available.

cs.CL↗

arXiv2021

Fast Dynamics in a Model Metallic Glass-forming Material

We investigate the fast $β$- and Johari-Goldstein (JG) $β$-relaxation processes, along with the elastic scattering response of glass-forming (GF) liquids and the Boson peak, in a simulated Al-Sm GF material exhibiting a fragile-strong (FS) transition. These dynamical processes are universal in 'ordinary' GF fluids and collectively describe their 'fast dynamics', and we find these relaxation processes also arise in a GF liquid exhibiting a fragile-strong transition. String-like particle motion, having both an irreversible and reversible nature ('stringlets') component, occurs in the fast-dynamics regime, corresponding to a ps timescale. String-like collective motion associated with localized unstable modes facilitate irreversible and intermittent particle 'jumping' events at long times associated with the JG $β$-relaxation process, while stringlets associated with localized stable modes and corresponding perfectly reversible atomic motion give rise to the Boson peak. To further clarify the origin of the Boson peak, we calculate the density of states for both the stringlet particles and the 'normal' particles and find that the stringlet particles give rise to a Boson peak while the normal atoms do not. The growth of stringlets upon heating ultimately also leads to the 'softening' of these excitations, and the Boson peak frequency and shear modulus drop in concert with this softening. The growth of string-like collective motion upon heating in the fast-dynamics regime is further shown to be responsible for the growth in the intensity of the fast relaxation process. Relaxation in cooled liquids clearly involves a hierarchy of relaxation processes acting on rather different time and spatial scales.

cond-mat.mtrl-sci↗

arXiv2020

Improving Target-side Lexical Transfer in Multilingual Neural Machine Translation

To improve the performance of Neural Machine Translation~(NMT) for low-resource languages~(LRL), one effective strategy is to leverage parallel data from a related high-resource language~(HRL). However, multilingual data has been found more beneficial for NMT models that translate from the LRL to a target language than the ones that translate into the LRLs. In this paper, we aim to improve the effectiveness of multilingual transfer for NMT models that translate \emph{into} the LRL, by designing a better decoder word embedding. Extending upon a general-purpose multilingual encoding method Soft Decoupled Encoding~\citep{SDE}, we propose DecSDE, an efficient character n-gram based embedding specifically designed for the NMT decoder. Our experiments show that DecSDE leads to consistent gains of up to 1.8 BLEU on translation from English to four different languages.

cs.CL↗

arXiv2020

Balancing Training for Multilingual Neural Machine Translation

When training multilingual machine translation (MT) models that can translate to/from multiple languages, we are faced with imbalanced training sets: some languages have much more training data than others. Standard practice is to up-sample less resourced languages to increase representation, and the degree of up-sampling has a large effect on the overall performance. In this paper, we propose a method that instead automatically learns how to weight training data through a data scorer that is optimized to maximize performance on all test languages. Experiments on two sets of languages under both one-to-many and many-to-one MT settings show our method not only consistently outperforms heuristic baselines in terms of average performance, but also offers flexible control over the performance of which languages are optimized.

cs.CL↗

arXiv2020

A Probabilistic Formulation of Unsupervised Text Style Transfer

We present a deep generative model for unsupervised text style transfer that unifies previously proposed non-generative techniques. Our probabilistic approach models non-parallel data from two domains as a partially observed parallel corpus. By hypothesizing a parallel latent sequence that generates each observed sequence, our model learns to transform sequences from one domain to another in a completely unsupervised fashion. In contrast with traditional generative sequence models (e.g. the HMM), our model makes few assumptions about the data it generates: it uses a recurrent language model as a prior and an encoder-decoder as a transduction distribution. While computation of marginal data likelihood is intractable in this model class, we show that amortized variational inference admits a practical surrogate. Further, by drawing connections between our variational objective and other recent unsupervised style transfer and machine translation techniques, we show how our probabilistic view can unify some known non-generative objectives such as backtranslation and adversarial loss. Finally, we demonstrate the effectiveness of our method on a wide range of unsupervised style transfer tasks, including sentiment transfer, formality transfer, word decipherment, author imitation, and related language translation. Across all style transfer tasks, our approach yields substantial gains over state-of-the-art non-generative baselines, including the state-of-the-art unsupervised machine translation techniques that our approach generalizes. Further, we conduct experiments on a standard unsupervised machine translation task and find that our unified approach matches the current state-of-the-art.

cs.CL↗

arXiv2019

Improving Conditioning in Context-Aware Sequence to Sequence Models

Neural sequence to sequence models are well established for applications which can be cast as mapping a single input sequence into a single output sequence. In this work, we focus on cases where generation is conditioned on both a short query and a long context, such as abstractive question answering or document-level translation. We modify the standard sequence-to-sequence approach to make better use of both the query and the context by expanding the conditioning mechanism to intertwine query and context attention. We also introduce a simple and efficient data augmentation method for the proposed model. Experiments on three different tasks show that both changes lead to consistent improvements.

cs.CL↗

arXiv2019

Domain Differential Adaptation for Neural Machine Translation

Neural networks are known to be data hungry and domain sensitive, but it is nearly impossible to obtain large quantities of labeled data for every domain we are interested in. This necessitates the use of domain adaptation strategies. One common strategy encourages generalization by aligning the global distribution statistics between source and target domains, but one drawback is that the statistics of different domains or tasks are inherently divergent, and smoothing over these differences can lead to sub-optimal performance. In this paper, we propose the framework of {\it Domain Differential Adaptation (DDA)}, where instead of smoothing over these differences we embrace them, directly modeling the difference between domains using models in a related task. We then use these learned domain differentials to adapt models for the target task accordingly. Experimental results on domain adaptation for neural machine translation demonstrate the effectiveness of this strategy, achieving consistent improvements over other alternative adaptation strategies in multiple experimental settings.

cs.CL↗

arXiv2019

compare-mt: A Tool for Holistic Comparison of Language Generation Systems

In this paper, we describe compare-mt, a tool for holistic analysis and comparison of the results of systems for language generation tasks such as machine translation. The main goal of the tool is to give the user a high-level and coherent view of the salient differences between systems that can then be used to guide further analysis or system improvement. It implements a number of tools to do so, such as analysis of accuracy of generation of particular types of words, bucketed histograms of sentence accuracies or counts based on salient characteristics, and extraction of characteristic $n$-grams for each system. It also has a number of advanced features such as use of linguistic labels, source side data, or comparison of log likelihoods for probabilistic models, and also aims to be easily extensible by users to new types of analysis. The code is available at https://github.com/neulab/compare-mt

cs.CL↗

arXiv2019

Target Conditioned Sampling: Optimizing Data Selection for Multilingual Neural Machine Translation

To improve low-resource Neural Machine Translation (NMT) with multilingual corpora, training on the most related high-resource language only is often more effective than using all data available (Neubig and Hu, 2018). However, it is possible that an intelligent data selection strategy can further improve low-resource NMT with data from other auxiliary languages. In this paper, we seek to construct a sampling distribution over all multilingual data, so that it minimizes the training loss of the low-resource language. Based on this formulation, we propose an efficient algorithm, Target Conditioned Sampling (TCS), which first samples a target sentence, and then conditionally samples its source sentence. Experiments show that TCS brings significant gains of up to 2 BLEU on three of four languages we test, with minimal training overhead.

cs.CL↗

arXiv2019

The ARIEL-CMU Systems for LoReHLT18

This paper describes the ARIEL-CMU submissions to the Low Resource Human Language Technologies (LoReHLT) 2018 evaluations for the tasks Machine Translation (MT), Entity Discovery and Linking (EDL), and detection of Situation Frames in Text and Speech (SF Text and Speech).

cs.CL↗

arXiv2019

Multilingual Neural Machine Translation With Soft Decoupled Encoding

Multilingual training of neural machine translation (NMT) systems has led to impressive accuracy improvements on low-resource languages. However, there are still significant challenges in efficiently learning word representations in the face of paucity of data. In this paper, we propose Soft Decoupled Encoding (SDE), a multilingual lexicon encoding framework specifically designed to share lexical-level information intelligently without requiring heuristic preprocessing such as pre-segmenting the data. SDE represents a word by its spelling through a character encoding, and its semantic meaning through a latent embedding space shared by all languages. Experiments on a standard dataset of four low-resource languages show consistent improvements over strong multilingual NMT baselines, with gains of up to 2 BLEU on one of the tested languages, achieving the new state-of-the-art on all four language pairs.

cs.CL↗

arXiv2019

Superionic UO2: A Model Anharmonic Crystalline Material

Crystalline materials at elevated temperatures and pressures can exhibit properties more reminiscent of simple liquids than ideal crystalline materials. Superionic crystalline materials having a liquid-like conductivity σ are particularly interesting for battery, fuel cell, and other energy applications, and we study UO2 as a prototypical superionic material since this material is widely studied given its commercial importance as reactor fuel. Using molecular dynamics, we first investigate basic thermodynamic and structural properties. We then quantify structural relaxation, dynamic heterogeneity, and average ions mobility. We find that the non-Arrhenius diffusion and structural relaxation time of this prototypical superionic material can be quantitatively described in terms of a generalized activated transport model ('string model') in which the activation energy varies in direct proportion to the average string length. Our transport data can also be described equally well by an Adam-Gibbs model in which the excess entropy density of the crystalline material is estimated from specific heat and thermal expansion data, consistent with the average scale of string-like collective motion scaling inversely with the excess entropy of the crystal. Strong differences in the temperature dependence of the interfacial mobility from non-ionic materials are observed, and we suggest that this difference is due to the relatively high cohesive interaction of ionic materials. In summary, the study of superionic UO2 provides insight into the role of cooperative motion in enhancing ion mobility in ionic materials and offers design principles for the development of new superionic materials for use in diverse energy applications.

cond-mat.mtrl-sci↗