arXiv subjects

Lei Lu

Publications and source records attributed to Lei Lu.

At least 37 records · Page 2Linked to original sources

arXiv2025

Understanding the Lyα Emission Observed by the Solar Disk Imager Aboard the Advanced Space-based Solar Observatory

The H I Lyman-alpha (Ly$α$) emission, with a wavelength of 1216 Å, is the brightest solar ultraviolet (UV) line. However, comprehensive observations of the Ly$α$ emission line across the full solar disk remain limited. As part of the ASO-S mission, the Solar Disk Imager (SDI) has successfully captured full-disk images in the Ly$α$ band. Gaussian fitting of SDI's spectral response function (SRF) yields a full width at half maximum (FWHM) of approximately 85 Å, which is significantly broader than the distance of Si III line at 1206 Å and the Ly$α$ line. Thus, the emission contribution of Si III to the SDI Ly$α$ passband needs to be considered. For flares, in practice, we calculated the integrated intensity ratio $I$(Si III)/$I$(Ly$α$) by analyzing spectral observations from the SOLSTICE instrument. It yields values between 1.7% and 14.6%. Empirically, the ratio is proportional to the SXR flux. Further analysis of spectral data from the SUMER instrument reveals that the ratio $I$(Si III)/$I$(Ly$α$) is approximately 0.5% for prominences, 0.7%--0.9% for the inner disk, and 1.4%--1.9% close to the limb. These findings suggest that $I$(Si III)/$I$(Ly$α$) is minimal for prominences and the inner disk, and the varying ratios across regions align with the center-to-limb variation of the Si III and Ly$α$ lines. Additionally, we compared Ly$α$ image intensity with 304 Å, 1600 Å, and 1700 Å observations from AIA, as well as H$α$ from CHASE, in multiple regions (a prominence region, two active regions, and a quiet region). A relatively higher correlation of about 85% is found between Ly$α$ and 304 Å in active regions, whereas in the quiet region and prominence, their correlation coefficients are about 55%.

astro-ph.SR↗

arXiv2025

Agile-Quant: Activation-Guided Quantization for Faster Inference of LLMs on the Edge

Large Language Models (LLMs) stand out for their impressive performance in intricate language modeling tasks. However, their demanding computational and memory needs pose obstacles for broad use on edge devices. Quantization is then introduced to boost LLMs' on-device efficiency. Recent works show that 8-bit or lower weight quantization is feasible with minimal impact on end-to-end task performance, while the activation is still not quantized. On the other hand, mainstream commodity edge devices still struggle to execute these sub-8-bit quantized networks effectively. In this paper, we propose Agile-Quant, an activation-guided quantization framework for popular Large Language Models (LLMs), and implement an end-to-end accelerator on multiple edge devices for faster inference. Considering the hardware profiling and activation analysis, we first introduce a basic activation quantization strategy to balance the trade-off of task performance and real inference speed. Then we leverage the activation-aware token pruning technique to reduce the outliers and the adverse impact on attentivity. Ultimately, we utilize the SIMD-based 4-bit multiplier and our efficient TRIP matrix multiplication to implement the accelerator for LLMs on the edge. We apply our framework on different scales of LLMs including LLaMA, OPT, and BLOOM with 4-bit or 8-bit for the activation and 4-bit for the weight quantization. Experiments show that Agile-Quant achieves simultaneous quantization of model weights and activations while maintaining task performance comparable to existing weight-only quantization methods. Moreover, in the 8- and 4-bit scenario, Agile-Quant achieves an on-device speedup of up to 2.55x compared to its FP16 counterparts across multiple edge devices, marking a pioneering advancement in this domain. Code: https://github.com/shawnricecake/agile-quant

cs.LG↗

arXiv2024

All-in-One Tuning and Structural Pruning for Domain-Specific LLMs

Existing pruning techniques for large language models (LLMs) targeting domain-specific applications typically follow a two-stage process: pruning the pretrained general-purpose LLMs and then fine-tuning the pruned LLMs on specific domains. However, the pruning decisions, derived from the pretrained weights, remain unchanged during fine-tuning, even if the weights have been updated. Therefore, such a combination of the pruning decisions and the finetuned weights may be suboptimal, leading to non-negligible performance degradation. To address these limitations, we propose ATP: All-in-One Tuning and Structural Pruning, a unified one-stage structural pruning and fine-tuning approach that dynamically identifies the current optimal substructure throughout the fine-tuning phase via a trainable pruning decision generator. Moreover, given the limited available data for domain-specific applications, Low-Rank Adaptation (LoRA) becomes a common technique to fine-tune the LLMs. In ATP, we introduce LoRA-aware forward and sparsity regularization to ensure that the substructures corresponding to the learned pruning decisions can be directly removed after the ATP process. ATP outperforms the state-of-the-art two-stage pruning methods on tasks in the legal and healthcare domains. More specifically, ATP recovers up to 88% and 91% performance of the dense model when pruning 40% parameters of LLaMA2-7B and LLaMA3-8B models, respectively.

cs.CL↗

arXiv2024

Task Adaptive Feature Distribution Based Network for Few-shot Fine-grained Target Classification

Metric-based few-shot fine-grained classification has shown promise due to its simplicity and efficiency. However, existing methods often overlook task-level special cases and struggle with accurate category description and irrelevant sample information. To tackle these, we propose TAFD-Net: a task adaptive feature distribution network. It features a task-adaptive component for embedding to capture task-level nuances, an asymmetric metric for calculating feature distribution similarities between query samples and support categories, and a contrastive measure strategy to boost performance. Extensive experiments have been conducted on three datasets and the experimental results show that our proposed algorithm outperforms recent incremental learning algorithms.

cs.CV↗

arXiv2024

Digital Avatars: Framework Development and Their Evaluation

We present a novel prompting strategy for artificial intelligence driven digital avatars. To better quantify how our prompting strategy affects anthropomorphic features like humor, authenticity, and favorability we present Crowd Vote - an adaptation of Crowd Score that allows for judges to elect a large language model (LLM) candidate over competitors answering the same or similar prompts. To visualize the responses of our LLM, and the effectiveness of our prompting strategy we propose an end-to-end framework for creating high-fidelity artificial intelligence (AI) driven digital avatars. This pipeline effectively captures an individual's essence for interaction and our streaming algorithm delivers a high-quality digital avatar with real-time audio-video streaming from server to mobile device. Both our visualization tool, and our Crowd Vote metrics demonstrate our AI driven digital avatars have state-of-the-art humor, authenticity, and favorability outperforming all competitors and baselines. In the case of our Donald Trump and Joe Biden avatars, their authenticity and favorability are rated higher than even their real-world equivalents.

cs.AI↗

arXiv2024

Inflight Performance and Calibrations of the Lyman-alpha Solar Telescope on board the Advanced Space-based Solar Observatory

The Lyman-alpha Solar Telescope (LST) on board the Advanced Space-based Solar Observatory (ASO-S) is the first payload to image the full solar disk and the solar corona in both white-light (WL) and ultraviolet (UV) H I Lya, extending up to 2.5 solar radii (Rs). Since the launch of the ASO-S on 9 October 2022, LST has captured various significant solar activities including flares, prominences, coronal mass ejections (CMEs). LST covers different passbands of 121.6 nm, 360 nm and 700 nm. The Lya Solar Disk Imager (SDI) has a field of view (FOV) of 38.4 arcmin and a spatial resolution of around 9.5 arcsec, while the White-Light Solar Telescope (WST) has a FOV of 38.43 arcmin and a spatial resolution of around 3.0 arcsec. The FOV of the Lya Solar Corona Imager (SCI) reaches 81.1 arcmin and its spatial resolution is 4.3 arcsec. The stray-light level in the 700 nm waveband is about 7.8e-6 MSB (mean solar brightness) at 1.1 Rs and 7.6e-7 MSB at 2.5 Rs, and in the Lya waveband it is around 4.3e-3 MSB at 1.1 Rs and 4.1e-4 MSB at 2.5 Rs. This article will detail the results from on-orbit tests and calibrations.

astro-ph.SR↗

arXiv2024

Association between a Failed Prominence Eruption and the Drainage of Mass from Another Prominence

Sympathetic eruptions of solar prominences have been studied for decades, however, it is usually difficult to identify their causal links. Here we present two failed prominence eruptions on 26 October 2022 and explore their connections. Using stereoscopic observations, the south prominence (PRO-S) erupts with untwisting motions, flare ribbons occur underneath, and new connections are formed during the eruption. The north prominence (PRO-N) rises up along with PRO-S, and its upper part disappears due to catastrophic mass draining along an elongated structure after PRO-S failed eruption. We suggest that the eruption of PRO-S initiates due to a kink instability, further rises up, and fails to erupt due to reconnection with surrounding fields. The elongated structure connecting PRO-N overlies PRO-S, which causes the rising up of PRO-N along with PRO-S and mass drainage after PRO-S eruption. This study suggests that a prominence may end its life through mass drainage forced by an eruption underneath.

astro-ph.SR↗

arXiv2024

Parameter effects on the total intensity of H I Lyα line for a modelled coronal mass ejection and its driven shock

The combination of the H I Lyα (121.6 nm) line formation mechanism with ultraviolet (UV) Lyα and white-light (WL) observations provides an effective method for determining the electron temperature of coronal mass ejections (CMEs). A key to ensuring the accuracy of this diagnostic technique is the precise calculation of theoretical Lyα intensities. This study performs a modelled CME and its driven shock via the 3D MHD simulation. We generate synthetic UV and WL images of the CME and shock to quantify the impact of different assumptions on theoretical Lyα intensities, such as the incident intensity of the Lyα line (Idisk), the geometric scattering function (p(θ)), and the kinetic temperature (Tn) assumed to be equal to the proton (Tp) or electron (Te) temperatures. By comparing differences of the Lyα intensities under these assumptions, we find that: (1) Using the uniform or Carrington maps of the disk Lyα emission underestimates the corona Lyα intensity (< 10%) compared to the synchronic map, except for a slight overestimate (< 4%) in the partial CME core. The Carrington map yields lower uncertainties than the uniform disk. (2) The geometric scattering process has a minor impact on the Lyα intensity, with a maximum relative uncertainty of < 5%. The Lyα intensity is underestimated for the most part but overestimated in the CME core. (3) Compared to the assumption Tn = Tp, using Tn = Te leads to more complex relative uncertainties in CME Lyα intensity. The CME core and void are both overestimated, with the maximum uncertainty in the core exceeding 50% and the void remaining below 35%. In the CME front, both over- and under-estimates exist with relative uncertainties of < 35%. The electron temperature assumption has a smaller impact on the shock, with an underestimated relative uncertainty of less than 20%.

astro-ph.SR↗

arXiv2024

CAMEL. II. A 3D Coronal Mass Ejection Catalog Based on Coronal Mass Ejection Automatic Detection with Deep Learning

Coronal mass ejections (CMEs) are major drivers of geomagnetic storms, which may cause severe space weather effects. Automating the detection, tracking, and three-dimensional (3D) reconstruction of CMEs is important for operational predictions of CME arrivals. The COR1 coronagraphs on board the Solar Terrestrial Relations Observatory spacecraft have facilitated extensive polarization observations, which are very suitable for the establishment of a 3D CME system. We have developed such a 3D system comprising four modules: classification, segmentation, tracking, and 3D reconstructions. We generalize our previously pretrained classification model to classify COR1 coronagraph images. Subsequently, as there are no publicly available CME segmentation data sets, we manually annotate the structural regions of CMEs using Large Angle and Spectrometric Coronagraph C2 observations. Leveraging transformer-based models, we achieve state-of-the-art results in CME segmentation. Furthermore, we improve the tracking algorithm to solve the difficult separation task of multiple CMEs. In the final module, tracking results, combined with the polarization ratio technique are used to develop the first single-view 3D CME catalog without requiring manual mask annotation. Our method provides higher precision in automatic 2D CME catalog and more reliable physical parameters of CMEs, including 3D propagation direction and speed. The aforementioned 3D CME system can be applied to any coronagraph data with the capability of polarization measurements.

astro-ph.SR↗

arXiv2024

Spectral and Imaging Observations of a C2.3 White-Light Flare from the Advanced Space-Based Solar Observatory (ASO-S) and the Chinese H$α$ Solar Explorer (CHASE)

Solar white-light flares are characterized by an enhancement in the optical continuum, which are usually large flares (say X- and M-class flares). Here we report a small C2.3 white-light flare (SOL2022-12-20T04:10) observed by the \emph{Advanced Space-based Solar Observatory} and the \emph{Chinese H$α$ Solar Explorer}. This flare exhibits an increase of $\approx$6.4\% in the photospheric Fe \textsc{i} line at 6569.2\,Å and {$\approx$3.2\%} in the nearby continuum. The continuum at 3600\,Å also shows an enhancement of $\approx$4.7\%. The white-light brightening kernels are mainly located at the flare ribbons and co-spatial with nonthermal hard X-ray sources, which implies that the enhanced white-light emissions are related to nonthermal electron-beam heating. At the brightening kernels, the Fe \textsc{i} line displays an absorption profile that has a good Gaussian shape, with a redshift up to $\approx$1.7 km s$^{-1}$, while the H$α$ line shows an emission profile though having a central reversal. The H$α$ line profile also shows a red or blue asymmetry caused by plasma flows with a velocity of several to tens of km s$^{-1}$. It is interesting to find that the H$α$ asymmetry is opposite at the conjugate footpoints. It is also found that the CHASE continuum increase seems to be related to the change of photospheric magnetic field. Our study provides comprehensive characteristics of a small white-light flare that help understand the energy release process of white-light flares.

astro-ph.SR↗

arXiv2024

HybridFlow: Infusing Continuity into Masked Codebook for Extreme Low-Bitrate Image Compression

This paper investigates the challenging problem of learned image compression (LIC) with extreme low bitrates. Previous LIC methods based on transmitting quantized continuous features often yield blurry and noisy reconstruction due to the severe quantization loss. While previous LIC methods based on learned codebooks that discretize visual space usually give poor-fidelity reconstruction due to the insufficient representation power of limited codewords in capturing faithful details. We propose a novel dual-stream framework, HyrbidFlow, which combines the continuous-feature-based and codebook-based streams to achieve both high perceptual quality and high fidelity under extreme low bitrates. The codebook-based stream benefits from the high-quality learned codebook priors to provide high quality and clarity in reconstructed images. The continuous feature stream targets at maintaining fidelity details. To achieve the ultra low bitrate, a masked token-based transformer is further proposed, where we only transmit a masked portion of codeword indices and recover the missing indices through token generation guided by information from the continuous feature stream. We also develop a bridging correction network to merge the two streams in pixel decoding for final image reconstruction, where the continuous stream features rectify biases of the codebook-based pixel decoder to impose reconstructed fidelity details. Experimental results demonstrate superior performance across several datasets under extremely low bitrates, compared with existing single-stream codebook-based or continuous-feature-based LIC methods.

eess.IV↗

arXiv2024

A Statistical Study of Solar White-Light Flares Observed by the White-light Solar Telescope of the Lyman-alpha Solar Telescope on the Advanced Space-based Solar Observatory (ASO-S/LST/WST) at 360 nm

Solar white-light flares (WLFs) are those accompanied by brightenings in the optical continuum or integrated light. The White-light Solar Telescope (WST), as an instrument of the Lyman-alpha Solar Telescope (LST) on the Advanced Space-based Solar Observatory (ASO-S), provides continuous solar full-disk images at 360 nm, which can be used to study WLFs. We analyze 205 major flares above M1.0 from October 2022 to May 2023 and identify 49 WLFs at 360 nm from WST observations, i.e. with an occurrence rate of 23.9%. The percentages of WLFs for M1 - M4 (31 out of 180), M5 - M9 (11 out of 18), and above X1 (7 for all) flares are 17.2%, 61.1%, and 100%, respectively, namely the larger the flares, the more likely they are WLFs at 360 nm. We further analyze 39 WLFs among the identified WLFs and investigate their properties such as white-light enhancement, duration, and brightening area. It is found that the relative enhancement of the white-light emission at 360 nm is mostly (>90%) less than 30% and the mean enhancement is 19.4%. The WLFs' duration at 360 nm is mostly (>80%) less than 20 minutes and its mean is 10.3 minutes. The brightening area at 360 nm is mostly (>75%) less than 500 arcsecond2 and the median value is 225. We find that there exist good correlations between the white-light enhancement/duration/area and the peak soft X-ray (SXR) flux of the flare, with correlation coefficients of 0.68, 0.58, and 0.80, respectively. In addition, the white-light emission in most WLFs peaks around the same time as the temporal derivative of SXR flux as well as the hard X-ray emission at 20 - 50 keV, indicative of Neupert effect. It is also found that the limb WLFs are more likely to have a greater enhancement, which is consistent with numerical simulations.

astro-ph.SR↗

arXiv2024

An Efficient and Scalable Clocking Assignment Algorithm for Multi-Threaded Multi-Phase Single Flux Quantum Circuits

A key distinguishing feature of single flux quantum (SFQ) circuits is that each logic gate is clocked. This feature forces the introduction of path-balancing flip-flops to ensure proper synchronization of inputs at each gate. This paper proposes a polynomial time complexity approximation algorithm for clocking assignments that minimizes the insertion of path balancing buffers for multi-threaded multi-phase clocking of SFQ circuits. Existing SFQ multi-phase clocking solutions have been shown to effectively reduce the number of required buffers inserted while maintaining high throughput, however, the associated clock assignment algorithms have exponential complexity and can have prohibitively long runtimes for large circuits, limiting the scalability of this approach. Our proposed algorithm is based on a linear program (LP) that leads to solutions that are experimentally on average within 5% of the optimum and helps accelerate convergence towards the optimal integer linear program (ILP) based solution. The improved LP and ILP runtimes permit multi-phase clocking schemes to scale to larger SFQ circuits than previous state of the art clocking assignment methods. We further extend the existing algorithm to support fanout sharing of the added buffers, saving, on average, an additional 10% of the inserted DFFs. Compared to traditional full path balancing (FPB) methods across 10 benchmarks, our enhanced LP saves 79.9%, 87.8%, and 91.2% of the inserted buffers for 2, 3, and 4 clock phases respectively. Finally, we extend this approach to the generation of circuits that completely mitigate potential hold-time violations at the cost of either adding on average less than 10% more buffers (for designs with 3 or more clock phases) or, more generally, adding a clock phase and thereby reducing throughput.

cs.ET↗

arXiv2024

Step-by-Step Canonical Quantum Gravity -- Part I: Ashtekar's New Variables

Canonical quantum gravity was first developed by Abhay Ashtekar, Lee Smolin, Carlo Rovelli and their collaborators in the late 1980s. It was a major breakthrough that successfully brought Einstein's theory of General Relativity (GR) into a Yang-Mills-type gauge theory. A new era of quantum gravity research has since started, and with decades of continued efforts from a relatively small community, the area now known as Loop Quantum Gravity (LQG) has flourished, making it a promising theory of quantum gravity. Due to its incredibly high level of complexity, many technical details were left out in introductory texts on LQG. In particular, resources that are appropriate to the undergraduate level are extremely limited. Consequently, there exists a huge gap between the knowledge base of an undergraduate physics major and the necessary readiness to carry out LQG research. In an effort to fill this gap, we aim to develop a pedagogical user guide that provides a step-by-step walk-through of canonical quantum gravity, without compromising necessary technical details. We hope that our attempt will bring more exposure to undergraduates on the exciting early developments of canonical quantum gravity, and provide them with the necessary foundation to explore active research fields such as black hole thermodynamics, Wheeler-DeWitt equation, and so on. This work will also serve as a solid base for anyone hoping to pursue further study in LQG at a higher level.

gr-qc↗

arXiv2023

A Two-Dimensional Deep Network for RF-based Drone Detection and Identification Towards Secure Coverage Extension

As drones become increasingly prevalent in human life, they also raises security concerns such as unauthorized access and control, as well as collisions and interference with manned aircraft. Therefore, ensuring the ability to accurately detect and identify between different drones holds significant implications for coverage extension. Assisted by machine learning, radio frequency (RF) detection can recognize the type and flight mode of drones based on the sampled drone signals. In this paper, we first utilize Short-Time Fourier. Transform (STFT) to extract two-dimensional features from the raw signals, which contain both time-domain and frequency-domain information. Then, we employ a Convolutional Neural Network (CNN) built with ResNet structure to achieve multi-class classifications. Our experimental results show that the proposed ResNet-STFT can achieve higher accuracy and faster convergence on the extended dataset. Additionally, it exhibits balanced performance compared to other baselines on the raw dataset.

eess.SP↗

arXiv2023

Semi-Supervised Learning for Multi-Label Cardiovascular Diseases Prediction:A Multi-Dataset Study

Electrocardiography (ECG) is a non-invasive tool for predicting cardiovascular diseases (CVDs). Current ECG-based diagnosis systems show promising performance owing to the rapid development of deep learning techniques. However, the label scarcity problem, the co-occurrence of multiple CVDs and the poor performance on unseen datasets greatly hinder the widespread application of deep learning-based models. Addressing them in a unified framework remains a significant challenge. To this end, we propose a multi-label semi-supervised model (ECGMatch) to recognize multiple CVDs simultaneously with limited supervision. In the ECGMatch, an ECGAugment module is developed for weak and strong ECG data augmentation, which generates diverse samples for model training. Subsequently, a hyperparameter-efficient framework with neighbor agreement modeling and knowledge distillation is designed for pseudo-label generation and refinement, which mitigates the label scarcity problem. Finally, a label correlation alignment module is proposed to capture the co-occurrence information of different CVDs within labeled samples and propagate this information to unlabeled samples. Extensive experiments on four datasets and three protocols demonstrate the effectiveness and stability of the proposed model, especially on unseen datasets. As such, this model can pave the way for diagnostic systems that achieve robust performance on multi-label CVDs prediction with limited supervision.

eess.SP↗

arXiv2023

Global energetics of solar powerful events on 6 September 2017

Solar flares and coronal mass ejections (CMEs) are thought to be the most powerful events on the Sun. They can release energy as high as 10^32 erg in tens of minutes,and could produce solar energetic particles (SEPs) in the interplanetary space. We explore global energy budgets of solar major eruptions on 6 September 2017, including the energy partition of a powerful solar flare, the energy budget of the accompanied CME and SEPs. In the wavelength range shortward of 222 nm, a major contribution of the flare radiated energy is in the soft X-ray (SXR) 0.1-7 nm domain. The flare energy radiated at wavelengths of Ly-alpha and middle ultraviolet is larger than that radiated in the extreme ultraviolet wavelength, but it is much less than that radiated in the SXR waveband. The total flare radiated energy could be comparable to the thermal and nonthermal energies. The energies carried by the major flare and its accompanied CME are roughly equal, and they are both powered by the magnetic free energy in the AR NOAA 12673. Moreover, the CME is efficient in accelerating SEPs, and that the prompt component (whether it comes from the solar flare or the CME) contributes only a negligible fraction.

astro-ph.SR↗

arXiv2022

Spatially Resolved Moving Radio Burst in Association with an EUV Wave

Coronal mass ejections (CMEs) are large clouds of magnetized plasma ejected from the Sun, and are often associated with acceleration of electrons that can result in radio emission via various mechanisms. However, the underlying mechanism relating the CMEs and particle acceleration still remains a subject of heated debate. Here, we report multi-instrument radio and extreme ultraviolet (EUV) imaging of a solar eruption event on 24 September 2011. We determine the emission mechanism of a moving radio burst, identify its three-dimensional (3D) location with respect to a rapidly expanding EUV wave, and find evidence for CME shocks that produce quasiperiodic acceleration of electron beams.

astro-ph.SR↗