arXiv subjects

Meng Chen

Publications and source records attributed to Meng Chen.

At least 91 records · Page 5Linked to original sources

arXiv2023

OTF: Optimal Transport based Fusion of Supervised and Self-Supervised Learning Models for Automatic Speech Recognition

Self-Supervised Learning (SSL) Automatic Speech Recognition (ASR) models have shown great promise over Supervised Learning (SL) ones in low-resource settings. However, the advantages of SSL are gradually weakened when the amount of labeled data increases in many industrial applications. To further improve the ASR performance when abundant labels are available, we first explore the potential of combining SL and SSL ASR models via analyzing their complementarity in recognition accuracy and optimization property. Then, we propose a novel Optimal Transport based Fusion (OTF) method for SL and SSL models without incurring extra computation cost in inference. Specifically, optimal transport is adopted to softly align the layer-wise weights to unify the two different networks into a single one. Experimental results on the public 1k-hour English LibriSpeech dataset and our in-house 2.6k-hour Chinese dataset show that OTF largely outperforms the individual models with lower error rates.

eess.AS↗

arXiv2023

Ordinal Outcome State-Space Models for Intensive Longitudinal Data

Intensive longitudinal (IL) data are increasingly prevalent in psychological science, coinciding with technological advancements that make it simple to deploy study designs such as daily diary and ecological momentary assessments. IL data are characterized by a rapid rate of data collection (1+ collections per day), over a period of time, allowing for the capture of the dynamics that underlie psychological and behavioral processes. One powerful framework for analyzing IL data is state-space modeling, where observed variables are considered measurements for underlying states (i.e., latent variables) that change together over time. However, state-space modeling has typically relied on continuous measurements, whereas psychological data often comes in the form of ordinal measurements such as Likert scale items. In this manuscript, we develop a general estimating approach for state-space models with ordinal measurements, specifically focusing on a graded response model for Likert scale items. We evaluate the performance of our model and estimator against that of the commonly used ``linear approximation'' model, which treats ordinal measurements as though they are continuous. We find that our model resulted in unbiased estimates of the state dynamics, while the linear approximation resulted in strongly biased estimates of the state dynamics

stat.ME↗

arXiv2023

Learning to Generate Poetic Chinese Landscape Painting with Calligraphy

In this paper, we present a novel system (denoted as Polaca) to generate poetic Chinese landscape painting with calligraphy. Unlike previous single image-to-image painting generation, Polaca takes the classic poetry as input and outputs the artistic landscape painting image with the corresponding calligraphy. It is equipped with three different modules to complete the whole piece of landscape painting artwork: the first one is a text-to-image module to generate landscape painting image, the second one is an image-to-image module to generate stylistic calligraphy image, and the third one is an image fusion module to fuse the two images into a whole piece of aesthetic artwork.

cs.CV↗

arXiv2023

A kernel-based least-squares collocation method for surface diffusion

There are plenty of applications and analysis for time-independent elliptic partial differential equations in the literature hinting at the benefits of overtesting by using more collocation conditions than the number of basis functions. Overtesting not only reduces the problem size, but is also known to be necessary for stability and convergence of widely used unsymmetric Kansa-type strong-form collocation methods. We consider kernel-based meshfree methods, which is a method of lines with collocation and overtesting spatially, for solving parabolic partial differential equations on surfaces without parametrization. In this paper, we extend the time-independent convergence theories for overtesting techniques to the parabolic equations on smooth and closed surfaces.

math.NA↗

arXiv2023

UFO2: A unified pre-training framework for online and offline speech recognition

In this paper, we propose a Unified pre-training Framework for Online and Offline (UFO2) Automatic Speech Recognition (ASR), which 1) simplifies the two separate training workflows for online and offline modes into one process, and 2) improves the Word Error Rate (WER) performance with limited utterance annotating. Specifically, we extend the conventional offline-mode Self-Supervised Learning (SSL)-based ASR approach to a unified manner, where the model training is conditioned on both the full-context and dynamic-chunked inputs. To enhance the pre-trained representation model, stop-gradient operation is applied to decouple the online-mode objectives to the quantizer. Moreover, in both the pre-training and the downstream fine-tuning stages, joint losses are proposed to train the unified model with full-weight sharing for the two modes. Experimental results on the LibriSpeech dataset show that UFO2 outperforms the SSL-based baseline method by 29.7% and 18.2% relative WER reduction in offline and online modes, respectively.

eess.AS↗

arXiv2022

Latent Evolution Model for Change Point Detection in Time-varying Networks

Graph-based change point detection (CPD) play an irreplaceable role in discovering anomalous graphs in the time-varying network. While several techniques have been proposed to detect change points by identifying whether there is a significant difference between the target network and successive previous ones, they neglect the natural evolution of the network. In practice, real-world graphs such as social networks, traffic networks, and rating networks are constantly evolving over time. Considering this problem, we treat the problem as a prediction task and propose a novel CPD method for dynamic graphs via a latent evolution model. Our method focuses on learning the low-dimensional representations of networks and capturing the evolving patterns of these learned latent representations simultaneously. After having the evolving patterns, a prediction of the target network can be achieved. Then, we can detect the change points by comparing the prediction and the actual network by leveraging a trade-off strategy, which balances the importance between the prediction network and the normal graph pattern extracted from previous networks. Intensive experiments conducted on both synthetic and real-world datasets show the effectiveness and superiority of our model.

cs.SI↗

arXiv2022

MNER-QG: An End-to-End MRC framework for Multimodal Named Entity Recognition with Query Grounding

Multimodal named entity recognition (MNER) is a critical step in information extraction, which aims to detect entity spans and classify them to corresponding entity types given a sentence-image pair. Existing methods either (1) obtain named entities with coarse-grained visual clues from attention mechanisms, or (2) first detect fine-grained visual regions with toolkits and then recognize named entities. However, they suffer from improper alignment between entity types and visual regions or error propagation in the two-stage manner, which finally imports irrelevant visual information into texts. In this paper, we propose a novel end-to-end framework named MNER-QG that can simultaneously perform MRC-based multimodal named entity recognition and query grounding. Specifically, with the assistance of queries, MNER-QG can provide prior knowledge of entity types and visual regions, and further enhance representations of both texts and images. To conduct the query grounding task, we provide manual annotations and weak supervisions that are obtained via training a highly flexible visual grounding model with transfer learning. We conduct extensive experiments on two public MNER datasets, Twitter2015 and Twitter2017. Experimental results show that MNER-QG outperforms the current state-of-the-art models on the MNER task, and also improves the query grounding performance.

cs.CV↗

arXiv2022

Privacy-Utility Balanced Voice De-Identification Using Adversarial Examples

Faced with the threat of identity leakage during voice data publishing, users are engaged in a privacy-utility dilemma when enjoying convenient voice services. Existing studies employ direct modification or text-based re-synthesis to de-identify users' voices, but resulting in inconsistent audibility in the presence of human participants. In this paper, we propose a voice de-identification system, which uses adversarial examples to balance the privacy and utility of voice services. Instead of typical additive examples inducing perceivable distortions, we design a novel convolutional adversarial example that modulates perturbations into real-world room impulse responses. Benefit from this, our system could preserve user identity from exposure by Automatic Speaker Identification (ASI) while remaining the voice perceptual quality for non-intrusive de-identification. Moreover, our system learns a compact speaker distribution through a conditional variational auto-encoder to sample diverse target embeddings on demand. Combining diverse target generation and input-specific perturbation construction, our system enables any-to-any identify transformation for adaptive de-identification. Experimental results show that our system could achieve 98% and 79% successful de-identification on mainstream ASIs and commercial systems with an objective Mel cepstral distortion of 4.31dB and a subjective mean opinion score of 4.48.

cs.SD↗

arXiv2022

Enabling Atomic Durability for Persistent Memory with Transiently Persistent CPU Cache

Persistent memory (pmem) products bring the persistence domain up to the memory level. Intel recently introduced the eADR feature that guarantees to flush data buffered in CPU cache to pmem on a power outage, thereby making the CPU cache a transient persistence domain. Researchers have explored how to enable the atomic durability for applications' in-pmem data. In this paper, we exploit the eADR-supported CPU cache to do so. A modified cache line, until written back to pmem, is a natural redo log copy of the in-pmem data. However, a write-back due to cache replacement or eADR on a crash overwrites the original copy. We accordingly develop Hercules, a hardware logging design for the transaction-level atomic durability, with supportive components installed in CPU cache, memory controller (MC), and pmem. When a transaction commits, Hercules commits on-chip its data staying in cache lines. For cache lines evicted before the commit, Hercules asks the MC to redirect and persist them into in-pmem log entries and commits them off-chip upon committing the transaction. Hercules lazily conducts pmem writes only for cache replacements at runtime. On a crash, Hercules saves metadata and data for active transactions into pmem for recovery. Experiments show that, by using CPU cache for both buffering and logging, Hercules yields much higher throughput and incurs significantly fewer pmem writes than state-of-the-art designs.

cs.AR↗

arXiv2022

On Power Control of Grid-Forming Converters: Modeling, Controllability, and Full-State Feedback Design

The popular single-input single-output control structures and classic design methods (e.g., root locus analysis) for the power control of grid-forming converters have limitations in applying to different line characteristics and providing favorable performance. This paper studies the grid-forming converter power loops from the perspective of multi-input multi-output systems. First, the error dynamics associated with power control loops (error-based state-space model) are derived while taking into account the natural dynamical coupling terms of the power converter models. Thereafter, the controllability Gramian of the grid-forming converter power loops is studied. Last, a full-state feedback control design using only the local measurements is applied. By this way, the eigenvalues of the system can be arbitrarily placed in the timescale of power loops based on predefined time-domain specifications. A step-by-step construction and design procedure of the power control of grid-forming converters is also given. The analysis and proposed method are verified by experimental results.

eess.SY↗

arXiv2022

Ultra-sensitive Flexible Sponge-Sensor Array for Muscle Activities Detection and Human Limb Motion Recognition

Human limb motion tracking and recognition plays an important role in medical rehabilitation training, lower limb assistance, prosthetics design for amputees, feedback control for assistive robots, etc. Lightweight wearable sensors, including inertial sensors, surface electromyography sensors, and flexible strain/pressure, are promising to become the next-generation human motion capture devices. Herein, we present a wireless wearable device consisting of a sixteen-channel flexible sponge-based pressure sensor array to recognize various human lower limb motions by detecting contours on the human skin caused by calf gastrocnemius muscle actions. Each sensing element is a round porous structure of thin carbon nanotube/polydimethylsiloxane nanocomposites with a diameter of 4 mm and thickness of about 400 μm. Ten human subjects were recruited to perform ten different lower limb motions while wearing the developed device. The motion classification result with the support vector machine method shows a macro-recall of about 97.3% for all ten motions tested. This work demonstrates a portable wearable muscle activity detection device with a lower limb motion recognition application, which can be potentially used in assistive robot control, healthcare, sports monitoring, etc.

eess.SP↗

arXiv2022

Multivariable Grid-Forming Converters with Direct States Control

A multi-input multi-output based grid-forming (MIMO-GFM) converter has been proposed using multivariable feedback control, which has been proven as a superior and robust system using low-order controllers. However, the original MIMO-GFM control is easily affected by the high-frequency components especially for the converter without inner cascaded voltage and current loops and when it is connected into a strong grid. This paper proposes an improved MIMO-GFM control method, where the frequency and internal voltage are chosen as state variables to be controlled directly. In this way, the impact of high-frequency components is eliminated without increasing the complexity of the control system. The H-infinity synthesis is used to tune the parameters to obtain an optimized performance. Experimental results verify the effectiveness of the proposed method.

eess.SY↗

arXiv2022

SCaLa: Supervised Contrastive Learning for End-to-End Speech Recognition

End-to-end Automatic Speech Recognition (ASR) models are usually trained to optimize the loss of the whole token sequence, while neglecting explicit phonemic-granularity supervision. This could result in recognition errors due to similar-phoneme confusion or phoneme reduction. To alleviate this problem, we propose a novel framework based on Supervised Contrastive Learning (SCaLa) to enhance phonemic representation learning for end-to-end ASR systems. Specifically, we extend the self-supervised Masked Contrastive Predictive Coding (MCPC) to a fully-supervised setting, where the supervision is applied in the following way. First, SCaLa masks variable-length encoder features according to phoneme boundaries given phoneme forced-alignment extracted from a pre-trained acoustic model; it then predicts the masked features via contrastive learning. The forced-alignment can provide phoneme labels to mitigate the noise introduced by positive-negative pairs in self-supervised MCPC. Experiments on reading and spontaneous speech datasets show that our proposed approach achieves 2.8 and 1.4 points Character Error Rate (CER) absolute reductions compared to the baseline, respectively.

eess.AS↗

arXiv2022

Power Control of Grid-Forming Converters Based on Full-State Feedback

The active and reactive power controllers of grid-forming converters are traditionally designed separately, which relies on the assumption of loop decoupling. This paper proposes a full-state feedback control for the power loops of grid-forming converters. First, the power loops are modeled considering their natural coupling, which, therefore, can apply to all kinds of line impedance, i.e., resistive, inductive, or complex. Then a full-state feedback control design is used. By this way, the eigenvalues of the system can be arbitrarily placed to any positions in the timescale of power loops. Therefore, the parameters can be directly chosen by the predefined specifications. A step-by-step parameters design procedure is also given in this paper. Experimental results verify the proposed method.

eess.SY↗

arXiv2022

A Bottom-Up End-User Intelligent Assistant Approach to Empower Gig Workers against AI Inequality

The growing inequality in gig work between workers and platforms has become a critical social issue as gig work plays an increasingly prominent role in the future of work. The AI inequality is caused by (1) the technology divide in who has access to AI technologies in gig work; and (2) the data divide in who owns the data in gig work leads to unfair working conditions, growing pay gap, neglect of workers' diverse preferences, and workers' lack of trust in the platforms. In this position paper, we argue that a bottom-up approach that empowers individual workers to access AI-enabled work planning support and share data among a group of workers through a network of end-user-programmable intelligent assistants is a practical way to bridge AI inequality in gig work under the current paradigm of privately owned platforms. This position paper articulates a set of research challenges, potential approaches, and community engagement opportunities, seeking to start a dialogue on this important research topic in the interdisciplinary CHIWORK community.

cs.HC↗

arXiv2022

Label Anchored Contrastive Learning for Language Understanding

Contrastive learning (CL) has achieved astonishing progress in computer vision, speech, and natural language processing fields recently with self-supervised learning. However, CL approach to the supervised setting is not fully explored, especially for the natural language understanding classification task. Intuitively, the class label itself has the intrinsic ability to perform hard positive/negative mining, which is crucial for CL. Motivated by this, we propose a novel label anchored contrastive learning approach (denoted as LaCon) for language understanding. Specifically, three contrastive objectives are devised, including a multi-head instance-centered contrastive loss (ICL), a label-centered contrastive loss (LCL), and a label embedding regularizer (LER). Our approach does not require any specialized network architecture or any extra data augmentation, thus it can be easily plugged into existing powerful pre-trained language models. Compared to the state-of-the-art baselines, LaCon obtains up to 4.1% improvement on the popular datasets of GLUE and CLUE benchmarks. Besides, LaCon also demonstrates significant advantages under the few-shot and data imbalance settings, which obtains up to 9.4% improvement on the FewGLUE and FewCLUE benchmarking tasks.

cs.CL↗

arXiv2022

SE-GAN: Skeleton Enhanced GAN-based Model for Brush Handwriting Font Generation

Previous works on font generation mainly focus on the standard print fonts where character's shape is stable and strokes are clearly separated. There is rare research on brush handwriting font generation, which involves holistic structure changes and complex strokes transfer. To address this issue, we propose a novel GAN-based image translation model by integrating the skeleton information. We first extract the skeleton from training images, then design an image encoder and a skeleton encoder to extract corresponding features. A self-attentive refined attention module is devised to guide the model to learn distinctive features between different domains. A skeleton discriminator is involved to first synthesize the skeleton image from the generated image with a pre-trained generator, then to judge its realness to the target one. We also contribute a large-scale brush handwriting font image dataset with six styles and 15,000 high-resolution images. Both quantitative and qualitative experimental results demonstrate the competitiveness of our proposed model.

cs.CV↗

arXiv2022

On minimal varieties growing from quasismooth weighted hypersurfaces

This paper concerns the construction of minimal varieties with small canonical volumes. The first part devotes to establishing an effective nefness criterion for the canonical divisor of a weighted blow-up over a weighted hypersurface, from which we construct plenty of new minimal $3$-folds including $59$ families of minimal $3$-folds of general type, several infinite series of minimal $3$-folds of Kodaira dimension $2$, $2$ families of minimal $3$-folds of general type on the Noether line, and $12$ families of minimal $3$-folds of general type near the Noether line. In the second part, we prove effective lower bounds of canonical volumes of minimal $n$-folds of general type with canonical dimension $n-1$ or $n-2$. Examples are provided to show that the theoretical lower bounds are optimal in dimension less than or equal to $5$ and nearly optimal in higher dimensions.

math.AG↗