arXiv subjects

Yong Ma

Publications and source records attributed to Yong Ma.

At least 37 records · Page 2Linked to original sources

arXiv2025

Multi-messenger dynamic imaging of laser-driven shocks in water using a plasma wakefield accelerator

Understanding dense matter hydrodynamics is critical for predicting plasma behavior in environments relevant to laser-driven inertial confinement fusion. Traditional diagnostic sources face limitations in brightness, spatiotemporal resolution, and inability to detect relevant electromagnetic fields. In this work, we present a dual-probe, multi-messenger laser wakefield accelerator platform combining ultrafast X-rays and relativistic electron beams at 1 Hz, to interrogate a free-flowing water target in vacuum, heated by an intense 200 ps laser pulse. This scheme enables high-repetition-rate tracking of the interaction evolution using both particle types. Betatron X-rays reveal a cylindrically symmetric shock compression morphology assisted by low-density vapor, resembling foam-layer-assisted fusion targets. The synchronized electron beam detects time-evolving electromagnetic fields, uncovering charge separation and ion species differentiation during plasma expansion - phenomena not captured by photons or hydrodynamic simulations. We show that combining both probes provides complementary insights spanning kinetic to hydrodynamic regimes, highlighting the need for hybrid physics models to accurately predict fusion-relevant plasma behavior

physics.plasm-ph↗

arXiv2025

HESSO: Towards Automatic Efficient and User Friendly Any Neural Network Training and Pruning

Structured pruning is one of the most popular approaches to effectively compress the heavy deep neural networks (DNNs) into compact sub-networks while retaining performance. The existing methods suffer from multi-stage procedures along with significant engineering efforts and human expertise. The Only-Train-Once (OTO) series has been recently proposed to resolve the many pain points by streamlining the workflow by automatically conducting (i) search space generation, (ii) structured sparse optimization, and (iii) sub-network construction. However, the built-in sparse optimizers in the OTO series, i.e., the Half-Space Projected Gradient (HSPG) family, have limitations that require hyper-parameter tuning and the implicit controls of the sparsity exploration, consequently requires intervening by human expertise. To address such limitations, we propose a Hybrid Efficient Structured Sparse Optimizer (HESSO). HESSO could automatically and efficiently train a DNN to produce a high-performing subnetwork. Meanwhile, it is almost tuning-free and enjoys user-friendly integration for generic training applications. To address another common issue of irreversible performance collapse observed in pruning DNNs, we further propose a Corrective Redundant Identification Cycle (CRIC) for reliably identifying indispensable structures. We numerically demonstrate the efficacy of HESSO and its enhanced version HESSO-CRIC on a variety of applications ranging from computer vision to natural language processing, including large language model. The numerical results showcase that HESSO can achieve competitive even superior performance to varying state-of-the-arts and support most DNN architectures. Meanwhile, CRIC can effectively prevent the irreversible performance collapse and further enhance the performance of HESSO on certain applications.

cs.LG↗

arXiv2025

State-of-the-Art HCI for Dementia Care: A Scoping Review of Recent Technological Advances

Dementia significantly impacts cognitive, behavioral, and functional abilities, creating challenges for both individuals and caregivers. Recent advancements in HCI have introduced innovative technological solutions to support people with dementia (PwD) and their caregivers. This scoping review systematically examines 32 recent publications from leading digital libraries, categorizing technological interventions into four key domains: Assistive and Smart Technology for Daily Life, Social Interaction and Communication, Well-being and Psychological Support, and Caregiver Support and Training. Our analysis highlights how emerging technologies are transforming dementia care. These technologies enhance quality of life by promoting independence, fostering social engagement, and providing emotional and cognitive support. However, the review also identifies critical gaps, particularly in addressing the needs of individuals with early-stage dementia and the lack of individualized support mechanisms. By emphasizing user-centered design, accessibility, and ethical considerations, this paper offers a structured roadmap for future research and practice in dementia care. It bridges the gap between technological innovation and the real-world needs of PwD and their caregivers, providing valuable insights for researchers, practitioners, and policymakers. This review not only synthesizes current advancements but also sets the stage for future HCI-driven innovations in dementia care, aiming to improve outcomes for an aging global population.

cs.HC↗

arXiv2025

Improving Acoustic Scene Classification in Low-Resource Conditions

Acoustic Scene Classification (ASC) identifies an environment based on an audio signal. This paper explores ASC in low-resource conditions and proposes a novel model, DS-FlexiNet, which combines depthwise separable convolutions from MobileNetV2 with ResNet-inspired residual connections for a balance of efficiency and accuracy. To address hardware limitations and device heterogeneity, DS-FlexiNet employs Quantization Aware Training (QAT) for model compression and data augmentation methods like Auto Device Impulse Response (ADIR) and Freq-MixStyle (FMS) to improve cross-device generalization. Knowledge Distillation (KD) from twelve teacher models further enhances performance on unseen devices. The architecture includes a custom Residual Normalization layer to handle domain differences across devices, and depthwise separable convolutions reduce computational overhead without sacrificing feature representation. Experimental results show that DS-FlexiNet excels in both adaptability and performance under resource-constrained conditions.

eess.AS↗

arXiv2025

Measuring User Experience Through Speech Analysis: Insights from HCI Interviews

User satisfaction plays a crucial role in user experience (UX) evaluation. Traditionally, UX measurements are based on subjective scales, such as questionnaires. However, these evaluations may suffer from subjective bias. In this paper, we explore the acoustic and prosodic features of speech to differentiate between positive and neutral UX during interactive sessions. By analyzing speech features such as root-mean-square (RMS), zero-crossing rate(ZCR), jitter, and shimmer, we identified significant differences between the positive and neutral user groups. In addition, social speech features such as activity and engagement also show notable variations between these groups. Our findings underscore the potential of speech analysis as an objective and reliable tool for UX measurement, contributing to more robust and bias-resistant evaluation methodologies. This work offers a novel approach to integrating speech features into UX evaluation and opens avenues for further research in HCI.

cs.HC↗

arXiv2025

Advancing User-Voice Interaction: Exploring Emotion-Aware Voice Assistants Through a Role-Swapping Approach

As voice assistants (VAs) become increasingly integrated into daily life, the need for emotion-aware systems that can recognize and respond appropriately to user emotions has grown. While significant progress has been made in speech emotion recognition (SER) and sentiment analysis, effectively addressing user emotions-particularly negative ones-remains a challenge. This study explores human emotional response strategies in VA interactions using a role-swapping approach, where participants regulate AI emotions rather than receiving pre-programmed responses. Through speech feature analysis and natural language processing (NLP), we examined acoustic and linguistic patterns across various emotional scenarios. Results show that participants favor neutral or positive emotional responses when engaging with negative emotional cues, highlighting a natural tendency toward emotional regulation and de-escalation. Key acoustic indicators such as root mean square (RMS), zero-crossing rate (ZCR), and jitter were identified as sensitive to emotional states, while sentiment polarity and lexical diversity (TTR) distinguished between positive and negative responses. These findings provide valuable insights for developing adaptive, context-aware VAs capable of delivering empathetic, culturally sensitive, and user-aligned responses. By understanding how humans naturally regulate emotions in AI interactions, this research contributes to the design of more intuitive and emotionally intelligent voice assistants, enhancing user trust and engagement in human-AI interactions.

cs.HC↗

arXiv2024

The 1st InterAI Workshop: Interactive AI for Human-centered Robotics

The workshop is affiliated with 33nd IEEE International Conference on Robot and Human Interactive Communication (RO-MAN 2024) August 26~30, 2023 / Pasadena, CA, USA. It is designed as a half-day event, extending over four hours from 9:00 to 12:30 PST time. It accommodates both in-person and virtual attendees (via Zoom), ensuring a flexible participation mode. The agenda is thoughtfully crafted to include a diverse range of sessions: two keynote speeches that promise to provide insightful perspectives, two dedicated paper presentation sessions, an interactive panel discussion to foster dialogue among experts which facilitates deeper dives into specific topics, and a 15-minute coffee break. The workshop website: https://sites.google.com/view/interaiworkshops/home.

cs.RO↗

arXiv2024

Vision Beyond Boundaries: An Initial Design Space of Domain-specific Large Vision Models in Human-robot Interaction

The emergence of large vision models (LVMs) is following in the footsteps of the recent prosperity of Large Language Models (LLMs) in following years. However, there's a noticeable gap in structured research applying LVMs to human-robot interaction (HRI), despite extensive evidence supporting the efficacy of vision models in enhancing interactions between humans and robots. Recognizing the vast and anticipated potential, we introduce an initial design space that incorporates domain-specific LVMs, chosen for their superior performance over normal models. We delve into three primary dimensions: HRI contexts, vision-based tasks, and specific domains. The empirical evaluation was implemented among 15 experts across six evaluated metrics, showcasing the primary efficacy in relevant decision-making scenarios. We explore the process of ideation and potential application scenarios, envisioning this design space as a foundational guideline for future HRI system design, emphasizing accurate domain alignment and model selection.

cs.HC↗

arXiv2024

Enhancing Source Code Classification Effectiveness via Prompt Learning Incorporating Knowledge Features

Researchers have investigated the potential of leveraging pre-trained language models, such as CodeBERT, to enhance source code-related tasks. Previous methodologies have relied on CodeBERT's '[CLS]' token as the embedding representation of input sequences for task performance, necessitating additional neural network layers to enhance feature representation, which in turn increases computational expenses. These approaches have also failed to fully leverage the comprehensive knowledge inherent within the source code and its associated text, potentially limiting classification efficacy. We propose CodeClassPrompt, a text classification technique that harnesses prompt learning to extract rich knowledge associated with input sequences from pre-trained models, thereby eliminating the need for additional layers and lowering computational costs. By applying an attention mechanism, we synthesize multi-layered knowledge into task-specific features, enhancing classification accuracy. Our comprehensive experimentation across four distinct source code-related tasks reveals that CodeClassPrompt achieves competitive performance while significantly reducing computational overhead.

cs.CL↗

arXiv2024

Rotation center identification based on geometric relationships for rotary motion deblurring

Non-blind rotary motion deblurring (RMD) aims to recover the latent clear image from a rotary motion blurred (RMB) image. The rotation center is a crucial input parameter in non-blind RMD methods. Existing methods directly estimate the rotation center from the RMB image. However they always suffer significant errors, and the performance of RMD is limited. For the assembled imaging systems, the position of the rotation center remains fixed. Leveraging this prior knowledge, we propose a geometric-based method for rotation center identification and analyze its error range. Furthermore, we construct a RMB imaging system. The experiment demonstrates that our method achieves less than 1-pixel error along a single axis (x-axis or y-axis). We utilize the constructed imaging system to capture real RMB images, and experimental results show that our method can help existing RMD approaches yield better RMD images.

cs.CV↗

arXiv2024

Bridging the Gap: Advancements in Technology to Support Dementia Care -- A Scoping Review

Dementia has serious consequences for the daily life of the person affected due to the decline in the their cognitive, behavioral and functional abilities. Caring for people living with dementia can be challenging and distressing. Innovative solutions are becoming essential to enrich the lives of those impacted and alleviate caregiver burdens. This scoping review, spanning literature from 2010 to July 2023 in the field of Human-Computer Interaction (HCI), offers a comprehensive look at how interactive technology contributes to dementia care. Emphasizing technology's role in addressing the unique needs of people with dementia (PwD) and their caregivers, this review encompasses assistive devices, mobile applications, sensors, and GPS tracking. Delving into challenges encountered in clinical and home-care settings, it succinctly outlines the influence of cutting-edge technologies, such as wearables, virtual reality, robots, and artificial intelligence, in supporting individuals with dementia and their caregivers. We categorize current dementia-related technologies into six groups based on their intended use and function: 1) daily life monitoring, 2) daily life support, 3) social interaction and communication, 4) well-being enhancement, 5) cognitive support, and 6) caregiver support.

cs.HC↗

arXiv2024

Mind Meets Robots: A Review of EEG-Based Brain-Robot Interaction Systems

Brain-robot interaction (BRI) empowers individuals to control (semi-)automated machines through their brain activity, either passively or actively. In the past decade, BRI systems have achieved remarkable success, predominantly harnessing electroencephalogram (EEG) signals as the central component. This paper offers an up-to-date and exhaustive examination of 87 curated studies published during the last five years (2018-2023), focusing on identifying the research landscape of EEG-based BRI systems. This review aims to consolidate and underscore methodologies, interaction modes, application contexts, system evaluation, existing challenges, and potential avenues for future investigations in this domain. Based on our analysis, we present a BRI system model with three entities: Brain, Robot, and Interaction, depicting the internal relationships of a BRI system. We especially investigate the essence and principles on interaction modes between human brains and robots, a domain that has not yet been identified anywhere. We then discuss these entities with different dimensions encompassed. Within this model, we scrutinize and classify current research, reveal insights, specify challenges, and provide recommendations for future research trajectories in this field. Meanwhile, we envision our findings offer a design space for future human-robot interaction (HRI) research, informing the creation of efficient BRI frameworks.

cs.RO↗

arXiv2024

A Novel Prompt-tuning Method: Incorporating Scenario-specific Concepts into a Verbalizer

The verbalizer, which serves to map label words to class labels, is an essential component of prompt-tuning. In this paper, we present a novel approach to constructing verbalizers. While existing methods for verbalizer construction mainly rely on augmenting and refining sets of synonyms or related words based on class names, this paradigm suffers from a narrow perspective and lack of abstraction, resulting in limited coverage and high bias in the label-word space. To address this issue, we propose a label-word construction process that incorporates scenario-specific concepts. Specifically, we extract rich concepts from task-specific scenarios as label-word candidates and then develop a novel cascade calibration module to refine the candidates into a set of label words for each class. We evaluate the effectiveness of our proposed approach through extensive experiments on {five} widely used datasets for zero-shot text classification. The results demonstrate that our method outperforms existing methods and achieves state-of-the-art results.

cs.CL↗

arXiv2023

Enhanced mechanical squeezing in an optomechanical system via backward stimulated Brillouin scattering

We investigate theoretically the enhancement of mechanical squeezing in a multimode optomechanical system by introducing a coherent phonon-photon interaction via the backward stimulated Brillouin scattering (BSBS) process. The coherent photon-phonon interaction where two optical modes couple to a Brillouin acoustic mode with a large decay rate provides an extra channel for the cooling of a Duffing mechanical oscillator. The squeezing degree and the robustness to the thermal noises of the Duffing mechanical mode can be enhanced greatly. When the Duffing nonlinearity is weak, the squeezing degree of the mechanical mode in the presence of BSBS can be improved more than one order of magnitude compared with the absence of BSBS. Our scheme may be extended to other quantum systems to study novel quantum effects.

quant-ph↗

arXiv2022

How Should Voice Assistants Deal With Users' Emotions?

There is a growing body of research in HCI on detecting the users' emotions. Once it is possible to detect users' emotions reliably, the next question is how an emotion-aware interface should react to the detected emotion. In a first step, we tried to find out how humans deal with the negative emotions of an avatar. The hope behind this approach was to identify human strategies, which we can then mimic in an emotion-aware voice assistant. We present a user study in which participants were confronted with an angry, sad, or frightened avatar. Their task was to make the avatar happy by talking to it. We recorded the voice signal and analyzed it. The results show that users predominantly reacted with neutral emotion. However, we also found gender differences, which opens a range of questions.

cs.HC↗

arXiv2021

An Innovative Transverse Emittance Cooling Technique using a Laser-Plasma Wiggler

We propose an innovative beam cooling scheme based on laser driven plasma wakefields to address the challenge of high luminosity generation for a future linear collider. For linear colliders, beam cooling is realised by means of damping rings equipped with wiggler magnets and accelerating cavities. This scheme ensures systematic reduction of phase space volume through synchrotron radiation emission whilst compensating for longitudinal momentum loss via an accelerating cavity. In this paper, the concept of a plasma wiggler and its effective model analogous to a magnetic wiggler are introduced; relation of plasma wiggler characteristics with damping properties are demonstrated; underpinning particle-in-cell simulations for laser propagation optimisation are presented. The oscillation of transverse wakefields and resulting sinusoidal probe beam trajectory are numerically demonstrated. The formation of an order of magnitude larger effective wiggler field compared to conventional wigglers is successfully illustrated. Potential damping ring designs on the basis of this novel plasma-based technology are presented and performance in terms of damping times and footprint was compared to an existing conventional damping ring design.

physics.acc-ph↗

arXiv2021

A Bright Source of Telecom Single Photons Based on Quantum Frequency Conversion

On-demand indistinguishable single photon sources are essential for quantum networking and communication. Semiconductor quantum dots are among the most promising candidates, but their typical emission wavelength renders them unsuitable for use in fibre networks. Here, we present quantum frequency conversion of near-infrared photons from a bright quantum dot to the telecommunication C-band, allowing integration with existing fibre architectures. We use a custom-built, tunable 2400 nm seed laser to convert single photons from 942 nm to 1550 nm in a difference frequency generation process. We achieve an end-to-end conversion efficiency of $\sim$35%, demonstrate count rates approaching 1 MHz at 1550 nm with $g^{\left(2\right)}\left(0\right) = 0.04$, and achieve Hong-Ou-Mandel visibilities of 60%. We expect this scheme to be preferable to quantum dot sources directly emitting at telecom wavelengths for fibre based quantum networking.

quant-ph↗

arXiv2021

Beyond optimization -- supervised learning applications in relativistic laser-plasma experiments

We explore the applications of machine learning techniques in relativistic laser-plasma experiments beyond optimization purposes. We predict the beam charge of electrons produced in a laser wakefield accelerator given the laser wavefront change caused by a deformable mirror. Machine learning enables feature analysis beyond merely searching for an optimal beam charge, showing that specific aberrations in the laser wavefront are favored in generating higher beam charges. Supervised learning models allow characterizing the measured data quality as well as recognizing irreproducible data and potential outliers. We also include virtual measurement errors in the experimental data to examine the model robustness under these conditions. This work demonstrates how machine learning methods can benefit data analysis and physics interpretation in a highly nonlinear problem of relativistic laser-plasma interaction.

physics.plasm-ph↗