arXiv subjects

Yimeng Wang

Publications and source records attributed to Yimeng Wang.

At least 37 records · Page 2Linked to original sources

arXiv2025

HiDream-I1: A High-Efficient Image Generative Foundation Model with Sparse Diffusion Transformer

Recent advancements in image generative foundation models have prioritized quality improvements but often at the cost of increased computational complexity and inference latency. To address this critical trade-off, we introduce HiDream-I1, a new open-source image generative foundation model with 17B parameters that achieves state-of-the-art image generation quality within seconds. HiDream-I1 is constructed with a new sparse Diffusion Transformer (DiT) structure. Specifically, it starts with a dual-stream decoupled design of sparse DiT with dynamic Mixture-of-Experts (MoE) architecture, in which two separate encoders are first involved to independently process image and text tokens. Then, a single-stream sparse DiT structure with dynamic MoE architecture is adopted to trigger multi-model interaction for image generation in a cost-efficient manner. To support flexiable accessibility with varied model capabilities, we provide HiDream-I1 in three variants: HiDream-I1-Full, HiDream-I1-Dev, and HiDream-I1-Fast. Furthermore, we go beyond the typical text-to-image generation and remould HiDream-I1 with additional image conditions to perform precise, instruction-based editing on given images, yielding a new instruction-based image editing model namely HiDream-E1. Ultimately, by integrating text-to-image generation and instruction-based image editing, HiDream-I1 evolves to form a comprehensive image agent (HiDream-A1) capable of fully interactive image creation and refinement. To accelerate multi-modal AIGC research, we have open-sourced all the codes and model weights of HiDream-I1-Full, HiDream-I1-Dev, HiDream-I1-Fast, HiDream-E1 through our project websites: https://github.com/HiDream-ai/HiDream-I1 and https://github.com/HiDream-ai/HiDream-E1. All features can be directly experienced via https://vivago.ai/studio.

cs.CV↗

arXiv2025

Will Your Next Pair Programming Partner Be Human? An Empirical Evaluation of Generative AI as a Collaborative Teammate in a Semester-Long Classroom Setting

Generative AI (GenAI), especially Large Language Models (LLMs), is rapidly reshaping both programming workflows and computer science education. Many programmers now incorporate GenAI tools into their workflows, including for collaborative coding tasks such as pair programming. While prior research has demonstrated the benefits of traditional pair programming and begun to explore GenAI-assisted coding, the role of LLM-based tools as collaborators in pair programming remains underexamined. In this work, we conducted a mixed-methods study with 39 undergraduate students to examine how GenAI influences collaboration, learning, and performance in pair programming. Specifically, students completed six in-class assignments under three conditions: Traditional Pair Programming (PP), Pair Programming with GenAI (PAI), and Solo Programming with GenAI (SAI). They used both LLM-based inline completion tools (e.g., GitHub Copilot) and LLM-based conversational tools (e.g., ChatGPT). Our results show that students in PAI achieved the highest assignment scores, whereas those in SAI attained the lowest. Additionally, students' attitudes toward LLMs' programming capabilities improved significantly after collaborating with LLM-based tools, and preferences were largely shaped by the perceived usefulness for completing assignments and learning programming skills, as well as the quality of collaboration. Our qualitative findings further reveal that while students appreciated LLM-based tools as valuable pair programming partners, they also identified limitations and had different expectations compared to human teammates. Our study provides one of the first empirical evaluations of GenAI as a pair programming collaborator through a comparison of three conditions (PP, PAI, and SAI). We also discuss the design implications and pedagogical considerations for future GenAI-assisted pair programming approaches.

cs.HC↗

arXiv2025

Evaluating an LLM-Powered Chatbot for Cognitive Restructuring: Insights from Mental Health Professionals

Recent advancements in large language models (LLMs) promise to expand mental health interventions by emulating therapeutic techniques, potentially easing barriers to care. Yet there is a lack of real-world empirical evidence evaluating the strengths and limitations of LLM-enabled psychotherapy interventions. In this work, we evaluate an LLM-powered chatbot, designed via prompt engineering to deliver cognitive restructuring (CR), with 19 users. Mental health professionals then examined the resulting conversation logs to uncover potential benefits and pitfalls. Our findings indicate that an LLM-based CR approach has the capability to adhere to core CR protocols, prompt Socratic questioning, and provide empathetic validation. However, issues of power imbalances, advice-giving, misunderstood cues, and excessive positivity reveal deeper challenges, including the potential to erode therapeutic rapport and ethical concerns. We also discuss design implications for leveraging LLMs in psychotherapy and underscore the importance of expert oversight to mitigate these concerns, which are critical steps toward safer, more effective AI-assisted interventions.

cs.HC↗

arXiv2025

Understanding Attitudes and Trust of Generative AI Chatbots for Social Anxiety Support

Social anxiety (SA) has become increasingly prevalent. Traditional coping strategies often face accessibility challenges. Generative AI (GenAI), known for their knowledgeable and conversational capabilities, are emerging as alternative tools for mental well-being. With the increased integration of GenAI, it is important to examine individuals' attitudes and trust in GenAI chatbots' support for SA. Through a mixed-method approach that involved surveys (n = 159) and interviews (n = 17), we found that individuals with severe symptoms tended to trust and embrace GenAI chatbots more readily, valuing their non-judgmental support and perceived emotional comprehension. However, those with milder symptoms prioritized technical reliability. We identified factors influencing trust, such as GenAI chatbots' ability to generate empathetic responses and its context-sensitive limitations, which were particularly important among individuals with SA. We also discuss the design implications and use of GenAI chatbots in fostering cognitive and emotional trust, with practical and design considerations.

cs.HC↗

arXiv2024

Emergent Symmetry and Valley Chern Insulator in Twisted Double-Bilayer Graphene

Theoretical calculations show that twisted double bilayer graphene (TDBG) under a transverse electric field develops a valley Chern number 2 at charge neutrality. Using thermodynamic and thermal activation measurements we report the experimental observation of a universal closing of the charge neutrality gap in the Hofstadter spectrum of TDBG at 1/2 magnetic flux per unit cell, in agreement with theoretical predictions for a valley Chern number 2 gap. Our theoretical analysis of the experimental data shows that the interaction energy, while larger than the flat-band bandwidth in TDBG near 1°does not alter the emergent valley symmetry or the single-particle band topology.

cond-mat.mes-hall↗

arXiv2024

Independently Tunable Flat Bands and Correlations in a Graphene Double Moiré System

We report on a double moiré system consisting of four graphene layers, where the top and bottom pairs form small-twist-angle bilayer graphene, and the middle interface has a large rotational mismatch. This system shows clear signatures of two sets of spatially separated flat bands associated with the top and bottom twisted bilayer graphene (TBG) subsystems, each independently tunable. Thermodynamic analysis reveals weak correlations between layers that allow the chemical potential to be measured as a function of carrier density for each constituent TBG. We find that correlated insulating states at integer number of electrons per moiré unit cell are most robust near magic angle, whereas gapped states at neutrality are more robust at larger twist angles.

cond-mat.mes-hall↗

arXiv2024

Integrated adaptive coherent LiDAR for 4D bionic vision

Light detection and ranging (LiDAR) is a ubiquitous tool to provide precise spatial awareness in various perception environments. A bionic LiDAR that can mimic human-like vision by adaptively gazing at selected regions of interest within a broad field of view is crucial to achieve high-resolution imaging in an energy-saving and cost-effective manner. However, current LiDARs based on stacking fixed-wavelength laser arrays and inertial scanning have not been able to achieve the desired dynamic focusing patterns and agile scalability simultaneously. Moreover, the ability to synchronously acquire multi-dimensional physical parameters, including distance, direction, Doppler, and color, through seamless fusion between multiple sensors, still remains elusive in LiDAR. Here, we overcome these limitations and demonstrate a bio-inspired frequency-modulated continuous wave (FMCW) LiDAR system with dynamic and scalable gazing capability. Our chip-scale LiDAR system is built using hybrid integrated photonic solutions, where a frequency-chirped external cavity laser provides broad spectral tunability, while on-chip electro-optic combs with elastic channel spacing allow customizable imaging granularity. Using the dynamic zoom-in capability and the coherent FMCW scheme, we achieve a state-of-the-art resolution of 0.012 degrees, providing up to 15 times the resolution of conventional 3D LiDAR sensors, with 115 equivalent scanning lines and 4D parallel imaging. We further demonstrate cooperative sensing between our adaptive coherent LiDAR and a camera to enable high-resolution color-enhanced machine vision.

physics.optics↗

arXiv2024

Meta-creatures: Developing an omnipotent hydrogel cell to construct bio-inspired systems

Due to current technological challenges, including the complexity of precise control, low long-term survival and success rates, difficulty in maintaining function over extended periods, and high energy consumption, the construction of life-like creatures with multilevel structures and varied physiological characteristics is a goal that has yet to be achieved1-3. Here, to create a parallel entity termed a meta-creature with similar functions and characteristics to those of natural organisms, we introduce a process for the transformation of an omnipotent hydrogel cell (OHC), which is inspired by totipotent stem cells and carries bio-inspired bioelectricity, into a meta-creature. We captured the electrical signals transmitted between the meta-nerve fibres and rat sciatic nerves, the physicochemical signals perceived by the meta-mouth, and the information exchanged between the meta-skin and the external environment. Notably, the meta-cardiovascular system, which is capable of exchanging matter and energy with external environments, exhibited similar electrocardiogram signals during testing in rabbits; these results indicated feedback with a biological system and the potential for ex vivo bioelectric remodelling. Finally, a meta-creature was designed and exhibited bio-inspired bioelectricity signals during the simulated outdoor flight. This work reveals new possibilities for constructing bio-inspired systems, thereby improving our understanding of bioelectricity and biomimicry.

physics.chem-ph↗

arXiv2024

"Stumbling-to-Fetters" mechanism and Virginia Creeper model in hydrogel for designing bionic cardiovascular system

Manufacturing hydrogels with identical electrochemical properties are typically riddled with unresolved inquiries and challenges. Here, we utilized ultra-light graphene flakes to trace the influence of convection phenomena during reactions on hydrogels' formation and structural non-uniformity, elucidating its mechanisms. Furthermore, we confirmed that an external electric field induced the orientation of functional groups of hydrogels along the direction of this field, revealing the mechanism of its influence on the structural non-uniformity and electrochemical properties of hydrogels. Additionally, we discovered that ion diffusion was "Stumbling-to-Fetters" by the functional groups on the polymer chains within the hydrogel, unveiling this mechanism and developing the Virginia Creeper (VC) model for hydrogels. We demonstrated the scalability and application of the VC model. Furthermore, we proposed a molecular-ion diffusion and current decay equation to describe the electrochemical properties of hydrogels. As an application of the VC model, we developed a bionic cardiovascular system and proved its potential to seamlessly interface with living organisms and generate bio-like bioelectricity. Our findings provide novel insights into triboelectricity and guidance for producing hydrogels with identical electrochemical properties, and offer a new pathway for bioelectric generation and the design of new hydrogel devices.

cond-mat.soft↗

arXiv2024

LMVD: A Large-Scale Multimodal Vlog Dataset for Depression Detection in the Wild

Depression can significantly impact many aspects of an individual's life, including their personal and social functioning, academic and work performance, and overall quality of life. Many researchers within the field of affective computing are adopting deep learning technology to explore potential patterns related to the detection of depression. However, because of subjects' privacy protection concerns, that data in this area is still scarce, presenting a challenge for the deep discriminative models used in detecting depression. To navigate these obstacles, a large-scale multimodal vlog dataset (LMVD), for depression recognition in the wild is built. In LMVD, which has 1823 samples with 214 hours of the 1475 participants captured from four multimedia platforms (Sina Weibo, Bilibili, Tiktok, and YouTube). A novel architecture termed MDDformer to learn the non-verbal behaviors of individuals is proposed. Extensive validations are performed on the LMVD dataset, demonstrating superior performance for depression detection. We anticipate that the LMVD will contribute a valuable function to the depression detection community. The data and code will released at the link: https://github.com/helang818/LMVD/.

cs.CV↗

arXiv2024

Evaluating the Effectiveness of LLMs in Introductory Computer Science Education: A Semester-Long Field Study

The integration of AI assistants, especially through the development of Large Language Models (LLMs), into computer science education has sparked significant debate. An emerging body of work has looked into using LLMs in education, but few have examined the impacts of LLMs on students in entry-level programming courses, particularly in real-world contexts and over extended periods. To address this research gap, we conducted a semester-long, between-subjects study with 50 students using CodeTutor, an LLM-powered assistant developed by our research team. Our study results show that students who used CodeTutor (the experimental group) achieved statistically significant improvements in their final scores compared to peers who did not use the tool (the control group). Within the experimental group, those without prior experience with LLM-powered tools demonstrated significantly greater performance gain than their counterparts. We also found that students expressed positive feedback regarding CodeTutor's capability, though they also had concerns about CodeTutor's limited role in developing critical thinking skills. Over the semester, students' agreement with CodeTutor's suggestions decreased, with a growing preference for support from traditional human teaching assistants. Our analysis further reveals that the quality of user prompts was significantly correlated with CodeTutor's response effectiveness. Building upon our results, we discuss the implications of our findings for integrating Generative AI literacy into curricula to foster critical thinking skills and turn to examining the temporal dynamics of user engagement with LLM-powered tools. We further discuss the discrepancy between the anticipated functions of tools and students' actual capabilities, which sheds light on the need for tailored strategies to improve educational outcomes.

cs.HC↗

arXiv2023

Aspect-Based Sentiment Analysis with Explicit Sentiment Augmentations

Aspect-based sentiment analysis (ABSA), a fine-grained sentiment classification task, has received much attention recently. Many works investigate sentiment information through opinion words, such as ''good'' and ''bad''. However, implicit sentiment widely exists in the ABSA dataset, which refers to the sentence containing no distinct opinion words but still expresses sentiment to the aspect term. To deal with implicit sentiment, this paper proposes an ABSA method that integrates explicit sentiment augmentations. And we propose an ABSA-specific augmentation method to create such augmentations. Specifically, we post-trains T5 by rule-based data. We employ Syntax Distance Weighting and Unlikelihood Contrastive Regularization in the training procedure to guide the model to generate an explicit sentiment. Meanwhile, we utilize the Constrained Beam Search to ensure the augmentation sentence contains the aspect terms. We test ABSA-ESA on two of the most popular benchmarks of ABSA. The results show that ABSA-ESA outperforms the SOTA baselines on implicit and explicit sentiment accuracy.

cs.CL↗

arXiv2023

Coherent control phase lag across doubly-excited atomic strontium resonances in an $ω-2ω$ interference scheme

Accurate calculations of phase lag associated with coherent control where an excited system decays into more than one product channel have recently been reported for atomic barium in PhysRevA.105.013113 (2022). The present study extends the calculations to make predictions of that observable for another alkaline earth atom, strontium, with a discussion of its spectrum and phase lag for energies between the $Sr^{+}$ $4d_{3/2}$ and $4d_{5/2}$ thresholds. We will explore the physics influenced by the electron correlations of strontium and the long-range Coulomb potential. The behavior of the phase lag cannot be simply addressed by a time delay analysis, although the latter is often used to address the prominent channel of resonance decay of doubly excited states.

physics.atom-ph↗

arXiv2023

Slow light silicon modulator beyond 110 GHz bandwidth

Silicon modulators are key components in silicon photonics to support the dense integration of electro-optic (EO) functional elements on a compact chip for various applications including high-speed data transmission, signal processing, and photonic computing. Despite numerous advances in promoting the operation speed of silicon modulators, a bandwidth ceiling of 67 GHz emerges in practices and becomes an obstacle to paving silicon photonics toward Tbps level data throughput on a single chip. Here, we theoretically propose and experimentally demonstrate a design strategy for silicon modulators by employing the slow light effect, which shatters the present bandwidth ceiling of silicon modulators and pushes its limit beyond 110 GHz in a small footprint. The proposed silicon modulator is built on a coupled-resonator optical waveguide (CROW) architecture, in which a set of Bragg gratings are appropriately cascaded to give rise to a slow light effect. By comprehensively balancing a series of merits including the group index, photon lifetime, electrical bandwidth, and losses, we found the modulators can benefit from the slow light for better modulation efficiency and compact size while remaining their bandwidth sufficiently high to support ultra-high-speed data transmission. Consequently, we realize a modulator with an EO bandwidth of 110 GHz in a length of 124 μm, and demonstrate a data rate beyond 110 Gbps by applying simple on-off keying modulation for a DSP-free operation. Our work proves that silicon modulators beyond 110 GHz are feasible, thus shedding light on the potentials of silicon photonics in ultra-high-bandwidth applications such as data communication, optical interconnection, and photonic machine learning.

physics.optics↗

arXiv2022

Attosecond Entangled Photons from Two-Photon Decay of Metastable Atoms: A Source for Attosecond Experiments and Beyond

We propose the generation of attosecond entangled bi-photons in the extreme-ultraviolet regime by two-photon decay of a metastable atomic state as a source similar to spontaneous parametric down-conversion photons. The 1s2s $^1S_0$ metastable state in helium decays to the ground state by emission of two energy-time entangled photons with a photon bandwidth equal to the total energy spacing of 20.62 eV. This results in a pair correlation time in the attosecond regime making these entangled photons a highly suitable source for attosecond pump-probe experiments. The bi-photon generation rate from a direct four photon excitation of helium at 240 nm is calculated and used to assess some feasible schemes to generate these bi-photons. Possible applications of entangled bi-photons in attosecond time scale experiments, and a discussion of their potential to reach the zeptosecond regime are presented.

quant-ph↗

arXiv2022

Emergence of Correlations in Alternating Twist Quadrilayer Graphene

Recently, alternating twist multilayer graphene (ATMG) has emerged as a family of moiré systems that share several fundamental properties with twisted bilayer graphene, and are expected to host similarly strong electron-electron interactions near the magic angle. Here, we study alternating twist quadrilayer graphene (ATQG) samples with twist angles of 1.96° and 1.52°, which are slightly removed from the magic angle of 1.68°. At the larger angle, we find signatures of correlated insulators only when the ATQG is hole doped, and no signatures of superconductivity, and for the smaller angle we find evidence of superconductivity, while signs of the correlated insulators weaken. Our results provide insight into the twist angle dependence of correlated phases in ATMG and shed light on the nature of correlations in the intermediate coupling regime at the edge of the magic angle range where dispersion and interaction are of the same order.

cond-mat.mes-hall↗

arXiv2022

Multichannel photoelectron phase lag across atomic barium autoionizing resonances

Phase lag associated with coherent control where an excited system decays into more than one product channel has been subjected to numerous investigations. Although previous theoretical studies have treated the phase lag across resonances in model calculations, quantitative agreement has never been achieved between the theoretical model and experimental measurements of phase lag from the $ω-2ω$ ionization of atomic barium \cite{PhysRevLett.98.053001,PhysRevA.76.053401}, suggesting that a toy model with phenomenological parameters is inadequate to describe the observed phase lag behavior. Here the phase lag is treated quantitatively from a multichannel coupling formulation, and our calculations based on multichannel quantum defect and $R$-matrix treatment achieves good agreement with the experimental observations. Our treatment also develops formulas to describe the effects of hyperfine depolarization on multiphoton ionization processes, and further, identifies resonances between $Ba^{+}$ $5d_{3/2}$ and $5d_{5/2}$ thresholds that have apparently never been experimentally observed and classified.

physics.atom-ph↗

arXiv2021

Adaptive F-FFT Demodulation for ICI Mitigation in Differential Underwater Acoustic OFDM Systems

This paper addresses the problem of frequency-domain inter-carrier interference (ICI) mitigation for differential orthogonal frequency-division multiplexing (OFDM) systems. The classical fractional fast Fourier transform (F-FFT), adopting the fixed sampling interval, would suffer from the limited accuracy of ICI mitigation and low adaptability in dynamic Doppler spread. To target the above challenges, we propose an adaptive fractional Fourier transform (A-FFT) demodulation method, in which an estimation algorithm based on the coordinate descent approach is designed to compute the fiducial frequency offset without increasing pilots. By means of compensating ICI at fractions of the fiducial frequency offset adapted to the time-varying Doppler shift, the A-FFT has the capability of tracking Doppler fluctuations over the underwater acoustic channels, thus extending the application range of frequency-domain ICI mitigation. Simulation results show that the A-FFT is significantly superior to the existing classical methods, the partial fast Fourier transform (P-FFT) and the F-FFT, for both medium and high Doppler factors and large carrier numbers in terms of the mean squared error (MSE). Numerically, the MSE of the A-FFT is reduced by $\bf{39.88\%-72.14\%}$ compared to that of the F-FFT with the input signal-to-noise ratio ranging from 10 dB to 30 dB at a Doppler factor of $\bf{2.5\times 10^{-4}}$ and a carrier number of 1024, while the P-FFT even cannot work well.

cs.IT↗