arXiv subjects

Yi Ding

Publications and source records attributed to Yi Ding.

At least 109 records · Page 6Linked to original sources

arXiv2025

FPGA-Optimized Hardware Accelerator for Fast Fourier Transform and Singular Value Decomposition in AI

This research introduces an FPGA-based hardware accelerator to optimize the Singular Value Decomposition (SVD) and Fast Fourier transform (FFT) operations in AI models. The proposed design aims to improve processing speed and reduce computational latency. Through experiments, we validate the performance benefits of the hardware accelerator and show how well it handles FFT and SVD operations. With its strong security and durability, the accelerator design achieves significant speedups over software implementations, thanks to its modules for data flow control, watermark embedding, FFT, and SVD.

cs.AR↗

arXiv2025

Discovery and inversion of the viscoelastic wave equation in inhomogeneous media

In scientific machine learning, the task of identifying partial differential equations accurately from sparse and noisy data poses a significant challenge. Current sparse regression methods may identify inaccurate equations on sparse and noisy datasets and are not suitable for varying coefficients. To address this issue, we propose a hybrid framework that combines two alternating direction optimization phases: discovery and embedding. The discovery phase employs current well-developed sparse regression techniques to preliminarily identify governing equations from observations. The embedding phase implements a recurrent convolutional neural network (RCNN), enabling efficient processes for time-space iterations involved in discretized forms of wave equation. The RCNN model further optimizes the imperfect sparse regression results to obtain more accurate functional terms and coefficients. Through alternating update of discovery-embedding phases, essential physical equations can be robustly identified from noisy and low-resolution measurements. To assess the performance of proposed framework, numerical experiments are conducted on various scenarios involving wave equation in elastic/viscoelastic and homogeneous/inhomogeneous media. The results demonstrate that the proposed method exhibits excellent robustness and accuracy, even when faced with high levels of noise and limited data availability in both spatial and temporal domains.

cs.LG↗

arXiv2025

Decoding Covert Speech from EEG Using a Functional Areas Spatio-Temporal Transformer

Covert speech involves imagining speaking without audible sound or any movements. Decoding covert speech from electroencephalogram (EEG) is challenging due to a limited understanding of neural pronunciation mapping and the low signal-to-noise ratio of the signal. In this study, we developed a large-scale multi-utterance speech EEG dataset from 57 right-handed native English-speaking subjects, each performing covert and overt speech tasks by repeating the same word in five utterances within a ten-second duration. Given the spatio-temporal nature of the neural activation process during speech pronunciation, we developed a Functional Areas Spatio-temporal Transformer (FAST), an effective framework for converting EEG signals into tokens and utilizing transformer architecture for sequence encoding. Our results reveal distinct and interpretable speech neural features by the visualization of FAST-generated activation maps across frontal and temporal brain regions with each word being covertly spoken, providing new insights into the discriminative features of the neural representation of covert speech. This is the first report of such a study, which provides interpretable evidence for speech decoding from EEG. The code for this work has been made public at https://github.com/Jiang-Muyun/FAST

eess.SP↗

arXiv2025

MamBEV: Enabling State Space Models to Learn Birds-Eye-View Representations

3D visual perception tasks, such as 3D detection from multi-camera images, are essential components of autonomous driving and assistance systems. However, designing computationally efficient methods remains a significant challenge. In this paper, we propose a Mamba-based framework called MamBEV, which learns unified Bird's Eye View (BEV) representations using linear spatio-temporal SSM-based attention. This approach supports multiple 3D perception tasks with significantly improved computational and memory efficiency. Furthermore, we introduce SSM based cross-attention, analogous to standard cross attention, where BEV query representations can interact with relevant image features. Extensive experiments demonstrate MamBEV's promising performance across diverse visual perception metrics, highlighting its advantages in input scaling efficiency compared to existing benchmark models.

cs.CV↗

arXiv2025

EmT: A Novel Transformer for Generalized Cross-subject EEG Emotion Recognition

Integrating prior knowledge of neurophysiology into neural network architecture enhances the performance of emotion decoding. While numerous techniques emphasize learning spatial and short-term temporal patterns, there has been limited emphasis on capturing the vital long-term contextual information associated with emotional cognitive processes. In order to address this discrepancy, we introduce a novel transformer model called emotion transformer (EmT). EmT is designed to excel in both generalized cross-subject EEG emotion classification and regression tasks. In EmT, EEG signals are transformed into a temporal graph format, creating a sequence of EEG feature graphs using a temporal graph construction module (TGC). A novel residual multi-view pyramid GCN module (RMPG) is then proposed to learn dynamic graph representations for each EEG feature graph within the series, and the learned representations of each graph are fused into one token. Furthermore, we design a temporal contextual transformer module (TCT) with two types of token mixers to learn the temporal contextual information. Finally, the task-specific output module (TSO) generates the desired outputs. Experiments on four publicly available datasets show that EmT achieves higher results than the baseline methods for both EEG emotion classification and regression tasks. The code is available at https://github.com/yi-ding-cs/EmT.

cs.LG↗

arXiv2025

Beyond Overfitting: Doubly Adaptive Dropout for Generalizable AU Detection

Facial Action Units (AUs) are essential for conveying psychological states and emotional expressions. While automatic AU detection systems leveraging deep learning have progressed, they often overfit to specific datasets and individual features, limiting their cross-domain applicability. To overcome these limitations, we propose a doubly adaptive dropout approach for cross-domain AU detection, which enhances the robustness of convolutional feature maps and spatial tokens against domain shifts. This approach includes a Channel Drop Unit (CD-Unit) and a Token Drop Unit (TD-Unit), which work together to reduce domain-specific noise at both the channel and token levels. The CD-Unit preserves domain-agnostic local patterns in feature maps, while the TD-Unit helps the model identify AU relationships generalizable across domains. An auxiliary domain classifier, integrated at each layer, guides the selective omission of domain-sensitive features. To prevent excessive feature dropout, a progressive training strategy is used, allowing for selective exclusion of sensitive features at any model layer. Our method consistently outperforms existing techniques in cross-domain AU detection, as demonstrated by extensive experimental evaluations. Visualizations of attention maps also highlight clear and meaningful patterns related to both individual and combined AUs, further validating the approach's effectiveness.

cs.CV↗

arXiv2025

Decoupled Doubly Contrastive Learning for Cross Domain Facial Action Unit Detection

Despite the impressive performance of current vision-based facial action unit (AU) detection approaches, they are heavily susceptible to the variations across different domains and the cross-domain AU detection methods are under-explored. In response to this challenge, we propose a decoupled doubly contrastive adaptation (D$^2$CA) approach to learn a purified AU representation that is semantically aligned for the source and target domains. Specifically, we decompose latent representations into AU-relevant and AU-irrelevant components, with the objective of exclusively facilitating adaptation within the AU-relevant subspace. To achieve the feature decoupling, D$^2$CA is trained to disentangle AU and domain factors by assessing the quality of synthesized faces in cross-domain scenarios when either AU or domain attributes are modified. To further strengthen feature decoupling, particularly in scenarios with limited AU data diversity, D$^2$CA employs a doubly contrastive learning mechanism comprising image and feature-level contrastive learning to ensure the quality of synthesized faces and mitigate feature ambiguities. This new framework leads to an automatically learned, dedicated separation of AU-relevant and domain-relevant factors, and it enables intuitive, scale-specific control of the cross-domain facial image synthesis. Extensive experiments demonstrate the efficacy of D$^2$CA in successfully decoupling AU and domain factors, yielding visually pleasing cross-domain synthesized facial images. Meanwhile, D$^2$CA consistently outperforms state-of-the-art cross-domain AU detection approaches, achieving an average F1 score improvement of 6\%-14\% across various cross-domain scenarios.

cs.CV↗

arXiv2025

Are We There Yet? A Measurement Study of Efficiency for LLM Applications on Mobile Devices

Recent advancements in large language models (LLMs) have prompted interest in deploying these models on mobile devices to enable new applications without relying on cloud connectivity. However, the efficiency constraints of deploying LLMs on resource-limited devices present significant challenges. In this paper, we conduct a comprehensive measurement study to evaluate the efficiency tradeoffs between mobile-based, edge-based, and cloud-based deployments for LLM applications. We implement AutoLife-Lite, a simplified LLM-based application that analyzes smartphone sensor data to infer user location and activity contexts. Our experiments reveal that: (1) Only small-size LLMs (<4B parameters) can run successfully on powerful mobile devices, though they exhibit quality limitations compared to larger models; (2) Model compression is effective in lower the hardware requirement, but may lead to significant performance degradation; (3) The latency to run LLMs on mobile devices with meaningful output is significant (>30 seconds), while cloud services demonstrate better time efficiency (<10 seconds); (4) Edge deployments offer intermediate tradeoffs between latency and model capabilities, with different results on CPU-based and GPU-based settings. These findings provide valuable insights for system designers on the current limitations and future directions for on-device LLM applications.

cs.PF↗

arXiv2025

ETA: Evaluating Then Aligning Safety of Vision Language Models at Inference Time

Vision Language Models (VLMs) have become essential backbones for multimodal intelligence, yet significant safety challenges limit their real-world application. While textual inputs are often effectively safeguarded, adversarial visual inputs can easily bypass VLM defense mechanisms. Existing defense methods are either resource-intensive, requiring substantial data and compute, or fail to simultaneously ensure safety and usefulness in responses. To address these limitations, we propose a novel two-phase inference-time alignment framework, Evaluating Then Aligning (ETA): 1) Evaluating input visual contents and output responses to establish a robust safety awareness in multimodal settings, and 2) Aligning unsafe behaviors at both shallow and deep levels by conditioning the VLMs' generative distribution with an interference prefix and performing sentence-level best-of-N to search the most harmless and helpful generation paths. Extensive experiments show that ETA outperforms baseline methods in terms of harmlessness, helpfulness, and efficiency, reducing the unsafe rate by 87.5% in cross-modality attacks and achieving 96.6% win-ties in GPT-4 helpfulness evaluation. The code is publicly available at https://github.com/DripNowhy/ETA.

cs.CV↗

arXiv2025

Exact Fit Attention in Node-Holistic Graph Convolutional Network for Improved EEG-Based Driver Fatigue Detection

EEG-based fatigue monitoring can effectively reduce the incidence of related traffic accidents. In the past decade, with the advancement of deep learning, convolutional neural networks (CNN) have been increasingly used for EEG signal processing. However, due to the data's non-Euclidean characteristics, existing CNNs may lose important spatial information from EEG, specifically channel correlation. Thus, we propose the node-holistic graph convolutional network (NHGNet), a model that uses graphic convolution to dynamically learn each channel's features. With exact fit attention optimization, the network captures inter-channel correlations through a trainable adjacency matrix. The interpretability is enhanced by revealing critical areas of brain activity and their interrelations in various mental states. In validations on two public datasets, NHGNet outperforms the SOTAs. Specifically, in the intra-subject, NHGNet improved detection accuracy by at least 2.34% and 3.42%, and in the inter-subjects, it improved by at least 2.09% and 15.06%. Visualization research on the model revealed that the central parietal area plays an important role in detecting fatigue levels, whereas the frontal and temporal lobes are essential for maintaining vigilance.

cs.LG↗

arXiv2025

Training-free Ultra Small Model for Universal Sparse Reconstruction in Compressed Sensing

Pre-trained large models attract widespread attention in recent years, but they face challenges in applications that require high interpretability or have limited resources, such as physical sensing, medical imaging, and bioinformatics. Compressed Sensing (CS) is a well-proved theory that drives many recent breakthroughs in these applications. However, as a typical under-determined linear system, CS suffers from excessively long sparse reconstruction times when using traditional iterative methods, particularly with large-scale data. Current AI methods like deep unfolding fail to substitute them because pre-trained models exhibit poor generality beyond their training conditions and dataset distributions, or lack interpretability. Instead of following the big model fervor, this paper proposes ultra-small artificial neural models called coefficients learning (CL), enabling training-free and rapid sparse reconstruction while perfectly inheriting the generality and interpretability of traditional iterative methods, bringing new feature of incorporating prior knowledges. In CL, a signal of length $n$ only needs a minimal of $n$ trainable parameters. A case study model called CLOMP is implemented for evaluation. Experiments are conducted on both synthetic and real one-dimensional and two-dimensional signals, demonstrating significant improvements in efficiency and accuracy. Compared to representative iterative methods, CLOMP improves efficiency by 100 to 1000 folds for large-scale data. Test results on eight diverse image datasets indicate that CLOMP improves structural similarity index by 292%, 98%, 45% for sampling rates of 0.1, 0.3, 0.5, respectively. We believe this method can truly usher CS reconstruction into the AI era, benefiting countless under-determined linear systems that rely on sparse solution.

cs.LG↗

arXiv2025

SelectiveFinetuning: Enhancing Transfer Learning in Sleep Staging through Selective Domain Alignment

In practical sleep stage classification, a key challenge is the variability of EEG data across different subjects and environments. Differences in physiology, age, health status, and recording conditions can lead to domain shifts between data. These domain shifts often result in decreased model accuracy and reliability, particularly when the model is applied to new data with characteristics different from those it was originally trained on, which is a typical manifestation of negative transfer. To address this, we propose SelectiveFinetuning in this paper. Our method utilizes a pretrained Multi Resolution Convolutional Neural Network (MRCNN) to extract EEG features, capturing the distinctive characteristics of different sleep stages. To mitigate the effect of domain shifts, we introduce a domain aligning mechanism that employs Earth Mover Distance (EMD) to evaluate and select source domain data closely matching the target domain. By finetuning the model with selective source data, our SelectiveFinetuning enhances the model's performance on target domain that exhibits domain shifts compared to the data used for training. Experimental results show that our method outperforms existing baselines, offering greater robustness and adaptability in practical scenarios where data distributions are often unpredictable.

eess.SP↗

arXiv2024

Towards Sustainable Large Language Model Serving

In this work, we study LLMs from a carbon emission perspective, addressing both operational and embodied emissions, and paving the way for sustainable LLM serving. We characterize the performance and energy of LLaMA with 1B, 3B, and 7B parameters using two Nvidia GPU types, a latest-generation RTX6000 Ada and an older-generation T4. We analytically model operational carbon emissions based on energy consumption and carbon intensities from three grid regions -- each representing a different energy source mix, and embodied carbon emissions based on chip area and memory size. Our characterization and modeling provide us with an in-depth understanding of the performance, energy, and carbon emissions of LLM serving. Our findings highlight the potential for optimizing sustainable LLM serving systems by considering both operational and embodied carbon emissions simultaneously.

cs.LG↗

arXiv2024

GreenLLM: Disaggregating Large Language Model Serving on Heterogeneous GPUs for Lower Carbon Emissions

LLMs have been widely adopted across many real-world applications. However, their widespread use comes with significant environmental costs due to their high computational intensity and resource demands. Specifically, this has driven the development of new generations of high-performing GPUs, exacerbating the problem of electronic waste and accelerating the premature disposal of devices. To address this problem, this paper focuses on reducing the carbon emissions of LLM serving by reusing older, low-performing GPUs. We present GreenLLM, an SLO-aware LLM serving framework designed to minimize carbon emissions by reusing older GPUs. GreenLLM builds on two identified use cases that disaggregate specific computations onto older GPUs, reducing carbon emissions while meeting performance goals. To deepen our understanding of the potential carbon savings from disaggregation, we also provide a theoretical analysis of its relationship with carbon intensity and GPU lifetime. Our evaluations show that GreenLLM reduces carbon emissions by up to 40.6% compared to running standard LLM serving on new GPU only, meeting latency SLOs for over 90% of requests across various applications, latency requirements, carbon intensities, and GPU lifetimes.

cs.AR↗

arXiv2024

Inducing Berry Curvature Dipole in Multilayer Graphene through Inhomogeneous Interlayer Sliding

Breaking lattice symmetry is crucial for generating a nonzero Berry curvature. While manipulating twisting angles between adjacent layers has successfully broken lattice symmetry through strain field and generated nonzero Berry curvature, interlayer sliding in principle offers a promising alternative route. However, realizing uniform interlayer sliding faces experimental challenges due to its energetic instability. In this work, we introduce an experimentally feasible method, using a corrugated substrate to induce an inhomogeneous but energetically more stable interlayer sliding in multilayer graphene. Our simulations demonstrate that inhomogeneous interlayer sliding produces a sizable Berry curvature dipole, which can be further tuned by varying the interlayer sliding distances and potential differences. The resulting Berry curvature dipole magnitude is remarkably up to 100 times greater than the maximum displacement involved in the inhomogeneous sliding. Our results highlight inhomogeneous interlayer sliding as a viable and effective method to induce a significant Berry curvature dipole in graphene systems and propose the experimentally feasible way to realize it.

cond-mat.mes-hall↗

arXiv2024

BiT-MamSleep: Bidirectional Temporal Mamba for EEG Sleep Staging

In this paper, we address the challenges in automatic sleep stage classification, particularly the high computational cost, inadequate modeling of bidirectional temporal dependencies, and class imbalance issues faced by Transformer-based models. To address these limitations, we propose BiT-MamSleep, a novel architecture that integrates the Triple-Resolution CNN (TRCNN) for efficient multi-scale feature extraction with the Bidirectional Mamba (BiMamba) mechanism, which models both short- and long-term temporal dependencies through bidirectional processing of EEG data. Additionally, BiT-MamSleep incorporates an Adaptive Feature Recalibration (AFR) module and a temporal enhancement block to dynamically refine feature importance, optimizing classification accuracy without increasing computational complexity. To further improve robustness, we apply optimization techniques such as Focal Loss and SMOTE to mitigate class imbalance. Extensive experiments on four public datasets demonstrate that BiT-MamSleep significantly outperforms state-of-the-art methods, particularly in handling long EEG sequences and addressing class imbalance, leading to more accurate and scalable sleep stage classification.

eess.SP↗

arXiv2024

Triad of Equivalence Theorems for the Radiant Intensity of Partially Coherent Beams on Scattering

By using Laplace's method for double integrals and the so-called beam condition obeyed by a partially coherent beamlike light field, we report the equivalence theory (ET) of partially coherent beams on scattering for the first time. We present the necessary and sufficient condition for the two scattered fields that have the same normalized radiant intensity distribution when Gaussian Schell-model beams whose effective beam widths are much greater than the effective transverse spectral coherence lengths are scattered by Gaussian Schell-model media. We find that the condition contain three implications, and each of them corresponds to a statement of an ET of radiant intensity in a scattering scenario, which exposes the concept of a previously unreported triad of ETs for the radiant intensity of partially coherent beams on scattering. We further find that the existing ET of plane waves on scattering, which only asserts that two scatterers with scattering potentials' correlations whose low-frequency antidiagonal spatial Fourier components are identical, essentially is merely the first member of our triad of ETs, while the other two hidden important members are completely ignored. Our findings are crucial for the inverse scattering problem since they contribute to avoid possible reconstruction errors in realistic situations, where the light field used to illuminate an unknown object is a partially coherent beam rather than an idealized plane wave.

physics.optics↗

arXiv2024

High Information Density and Low Coverage Data Storage in DNA with Efficient Channel Coding Schemes

DNA-based data storage has been attracting significant attention due to its extremely high data storage density, low power consumption, and long duration compared to conventional data storage media. Despite the recent advancements in DNA data storage technology, significant challenges remain. In particular, various types of errors can occur during the processes of DNA synthesis, storage, and sequencing, including substitution errors, insertion errors, and deletion errors. Furthermore, the entire oligo may be lost. In this work, we report a DNA-based data storage architecture that incorporates efficient channel coding schemes, including different types of error-correcting codes (ECCs) and constrained codes, for both the inner coding and outer coding for the DNA data storage channel. We also carried out large scale experiments to validate our proposed DNA-based data storage architecture. Specifically, 1.61 and 1.69 MB data were encoded into 30,000 oligos each, with information densities of 1.731 and 1.815, respectively. It has been found that the stored information can be fully recovered without any error at average coverages of 4.5 and 6.0, respectively. This experiment achieved the highest net information density and lowest coverage among existing DNA-based data storage experiments (with standard DNA), with data recovery rates and coverage approaching theoretical optima.

cs.IT↗