arXiv subjects

Kai Li

Publications and source records attributed to Kai Li.

At least 217 records · Page 12Linked to original sources

arXiv2025

Optical Flares Detected on a Contact Binary: The First Photometric and Spectroscopic Analysis of a Long-period Low Mass Ratio Contact Binary HAT 307-0007476

This paper presents the photometric and spectroscopic analysis of a long-period totally eclipsing contact binary (HAT 307-0007476) for the first time. This system is a low mass ratio ($q\sim0.114$) and medium contact binary ($f\sim37.1\%$). Two flare events were detected in multiple bands observations in December 2022. The interval between the two flare events is 4 days. The average duration of these two flares is about 2289s. Both the two flares achieve the energy levels of superflares. The excess emission of the H$_α$ line in the LAMOST spectra of this object was analyzed, indicating its chromospheric activity. The $O-C$ diagram showed a long-term orbital period increase, which is due to the mass transfer between the two component stars. We conclude that HAT 307-0007476 is currently in a stable region based on both $J_{spin}/J_{orb}$ and the comparison between the instability parameters and its current values.

astro-ph.SR↗

arXiv2025

Knowledge Transfer and Domain Adaptation for Fine-Grained Remote Sensing Image Segmentation

Fine-grained remote sensing image segmentation is essential for accurately identifying detailed objects in remote sensing images. Recently, vision transformer models (VTMs) pre-trained on large-scale datasets have demonstrated strong zero-shot generalization. However, directly applying them to specific tasks may lead to domain shift. We introduce a novel end-to-end learning paradigm combining knowledge guidance with domain refinement to enhance performance. We present two key components: the Feature Alignment Module (FAM) and the Feature Modulation Module (FMM). FAM aligns features from a CNN-based backbone with those from the pretrained VTM's encoder using channel transformation and spatial interpolation, and transfers knowledge via KL divergence and L2 normalization constraint. FMM further adapts the knowledge to the specific domain to address domain shift. We also introduce a fine-grained grass segmentation dataset and demonstrate, through experiments on two datasets, that our method achieves a significant improvement of 2.57 mIoU on the grass dataset and 3.73 mIoU on the cloud dataset. The results highlight the potential of combining knowledge transfer and domain adaptation to overcome domain-related challenges and data limitations. The project page is available at https://xavierjiezou.github.io/KTDA/.

cs.CV↗

arXiv2025

Apollo: Band-sequence Modeling for High-Quality Audio Restoration

Audio restoration has become increasingly significant in modern society, not only due to the demand for high-quality auditory experiences enabled by advanced playback devices, but also because the growing capabilities of generative audio models necessitate high-fidelity audio. Typically, audio restoration is defined as a task of predicting undistorted audio from damaged input, often trained using a GAN framework to balance perception and distortion. Since audio degradation is primarily concentrated in mid- and high-frequency ranges, especially due to codecs, a key challenge lies in designing a generator capable of preserving low-frequency information while accurately reconstructing high-quality mid- and high-frequency content. Inspired by recent advancements in high-sample-rate music separation, speech enhancement, and audio codec models, we propose Apollo, a generative model designed for high-sample-rate audio restoration. Apollo employs an explicit frequency band split module to model the relationships between different frequency bands, allowing for more coherent and higher-quality restored audio. Evaluated on the MUSDB18-HQ and MoisesDB datasets, Apollo consistently outperforms existing SR-GAN models across various bit rates and music genres, particularly excelling in complex scenarios involving mixtures of multiple instruments and vocals. Apollo significantly improves music restoration quality while maintaining computational efficiency. The source code for Apollo is publicly available at https://github.com/JusperLee/Apollo.

cs.SD↗

arXiv2024

Minute-cadence Observations of the LAMOST Fields with the TMTS: VI. Absolute Physical Parameters of Contact Binaries

With the development of wide-field surveys, a large amount of data on short-period W UMa contact binaries have been obtained. Continuous and uninterrupted light curves as well as high-resolution spectroscopic data are crucial in determining the absolute physical parameters. Targets with both TMTS light curves and LAMOST medium-resolution spectra were selected. The absolute physical parameters were inferred with the W-D code for ten systems, all of them are W-type shallow or medium contact binaries. The O'Connell effect observed in the light curves can be explained by adding a spot on the primary or secondary component in the models. According to O-C analysis, the orbital periods exhibit a long-term increasing or decreasing trend, amongst which J0132, J1300, and J1402 show periodic variations that may be attributed to the presence of a third body or magnetic activity cycles. Spectral subtraction analysis revealed that the equivalent width of H$α$ indicates strong magnetic activity in J0047, J0305, J0638, and J1402. Among the 10 selected binary systems, except for J0132 and J0913, the more massive components are found to be main-sequence stars while the less massive components have evolved off the main sequence. In J0132, both components are in the main sequence, whereas both components of J0913 lie above the terminal-age main sequence. Based on the relationship between orbital angular momentum and total mass for these two systems, as well as their low fill-out factors, it is possible that these two systems are newly formed contact binaries, having recently evolved from the detached configuration.

astro-ph.SR↗

arXiv2024

HES-UNet: A U-Net for Hepatic Echinococcosis Lesion Segmentation

Hepatic echinococcosis (HE) is a prevalent disease in economically underdeveloped pastoral areas, where adequate medical resources are usually lacking. Existing methods often ignore multi-scale feature fusion or focus only on feature fusion between adjacent levels, which may lead to insufficient feature fusion. To address these issues, we propose HES-UNet, an efficient and accurate model for HE lesion segmentation. This model combines convolutional layers and attention modules to capture local and global features. During downsampling, the multi-directional downsampling block (MDB) is employed to integrate high-frequency and low-frequency features, effectively extracting image details. The multi-scale aggregation block (MAB) aggregates multi-scale feature information. In contrast, the multi-scale upsampling Block (MUB) learns highly abstract features and supplies this information to the skip connection module to fuse multi-scale features. Due to the distinct regional characteristics of HE, there is currently no publicly available high-quality dataset for training our model. We collected CT slice data from 268 patients at a certain hospital to train and evaluate the model. The experimental results show that HES-UNet achieves state-of-the-art performance on our dataset, achieving an overall Dice Similarity Coefficient (DSC) of 89.21%, which is 1.09% higher than that of TransUNet. The project page is available at https://chenjiayan-qhu.github.io/HES-UNet-page.

eess.IV↗

arXiv2024

InstantSwap: Fast Customized Concept Swapping across Sharp Shape Differences

Recent advances in Customized Concept Swapping (CCS) enable a text-to-image model to swap a concept in the source image with a customized target concept. However, the existing methods still face the challenges of inconsistency and inefficiency. They struggle to maintain consistency in both the foreground and background during concept swapping, especially when the shape difference is large between objects. Additionally, they either require time-consuming training processes or involve redundant calculations during inference. To tackle these issues, we introduce InstantSwap, a new CCS method that aims to handle sharp shape disparity at speed. Specifically, we first extract the bbox of the object in the source image automatically based on attention map analysis and leverage the bbox to achieve both foreground and background consistency. For background consistency, we remove the gradient outside the bbox during the swapping process so that the background is free from being modified. For foreground consistency, we employ a cross-attention mechanism to inject semantic information into both source and target concepts inside the box. This helps learn semantic-enhanced representations that encourage the swapping process to focus on the foreground objects. To improve swapping speed, we avoid computing gradients at each timestep but instead calculate them periodically to reduce the number of forward passes, which improves efficiency a lot with a little sacrifice on performance. Finally, we establish a benchmark dataset to facilitate comprehensive evaluation. Extensive evaluations demonstrate the superiority and versatility of InstantSwap. Project Page: https://instantswap.github.io/

cs.CV↗

arXiv2024

Photometric and Spectroscopic Investigations of Three Large Amplitude Contact Binaries

We performed photometric and spectroscopic studies of three large amplitude contact binaries, NSVS 2418361, ATLAS J057.1170+31.2384 and NSVS 7377875. The amplitudes of three systems' light curves are more than 0.7 magnitude. We analyzed the light curves using Wilson-Devinney code to yield physical parameters. The photometric solutions suggested that NSVS 7377875 belongs to an A-subtype contact binary, while the others are classified as W-subtype ones. Furthermore, the mass ratio of NSVS 7377875 is higher than 0.72, so it belongs to H-subtype contact binaries. Since their light curves have unequal height at two maxima which is called O'Connell effect, a dark spot on the primary component for each target was required to get a better fit of light curves. The orbital period investigation shows that the period of NSVS 2418361 is increasing, indicating a mass transfer from the less massive component to the more massive one, while the other targets exhibit no long-term variation. Our spectral subtraction analysis of LAMOST spectra revealed excess emissions in the $H_α$ line, indicating chromospheric activity in all the three targets. The Gaia distance was applied to estimate the absolute parameters of the three targets, and we obtained their evolutionary state. The relationships between the energy transfer parameter of 76 H-subtype contact binaries and their bolometric luminosity ratios, as well as their contact degree, were presented. We discovered that H-subtype systems have less efficient energy transfer rate, which is corresponding to the conclusion proposed by Csizmadia \& Klagyivik.

astro-ph.SR↗

arXiv2024

Search for and analysis of eclipsing binaries in the LAMOST Medium-Resolution Survey field. I. RA: $\textbf{23}^h$$\textbf{01}^m$$\textbf{51}^s$, Dec: +34$^\circ$36$^\prime$45$^{\prime \prime}$

Eclipsing binaries (EBs) play an important astrophysical role in studying stellar properties and evolution. By analyzing photometric data in the LAMOST Medium-Resolution Survey field, RA: $23^h$$01^m$$51.00^s$, Dec: +34$^\circ$36$^\prime$45$^{\prime \prime}$, 48 EBs are detected and 2 are newly discovered. This specific field has been observed 52 times by the LAMOST Medium-Resolution Survey DR 9, which facilitates a comprehensive analysis of the EBs. For EBs with LAMOST medium-resolution spectra, radial velocity curves were obtained, and their precise orbital parameters were determined by simultaneously analyzing photometric light curves and radial velocity curves. For the other EBs with only photometric light curves, we used the q-search or the temperature ratio method to determine their initial mass ratios and then determined the orbital parameters. It is found that 15 EBs belong to detached systems, 1 to semi-detached systems, and 32 to contact systems. Based on the O-C analysis for 26 EBs with sufficient eclipsing times, we found a long-term decrease in the orbital period of 11 EBs and a continuous increase of 5 EBs, which are due to the material transfer between the two components. The O-C curve of 1 EB shows a distinct periodic variation, which is caused by the light travel time effect, and the third body is likely to be a black hole. By applying the spectral subtraction method to 13 EBs with LAMOST medium-resolution spectra, 10 systems exhibit distinct H$α$ emission lines, in which 1 system exhibits double-peaked lines near phases 0.25 and 0.75, implying strong chromospheric activity. In the mass-luminosities and mass-radius distributions, most of the more massive components are less evolved than the less massive ones.

astro-ph.SR↗

arXiv2024

Adapting Vision Foundation Models for Robust Cloud Segmentation in Remote Sensing Images

Cloud segmentation is a critical challenge in remote sensing image interpretation, as its accuracy directly impacts the effectiveness of subsequent data processing and analysis. Recently, vision foundation models (VFM) have demonstrated powerful generalization capabilities across various visual tasks. In this paper, we present a parameter-efficient adaptive approach, termed Cloud-Adapter, designed to enhance the accuracy and robustness of cloud segmentation. Our method leverages a VFM pretrained on general domain data, which remains frozen, eliminating the need for additional training. Cloud-Adapter incorporates a lightweight spatial perception module that initially utilizes a convolutional neural network (ConvNet) to extract dense spatial representations. These multi-scale features are then aggregated and serve as contextual inputs to an adapting module, which modulates the frozen transformer layers within the VFM. Experimental results demonstrate that the Cloud-Adapter approach, utilizing only 0.6% of the trainable parameters of the frozen backbone, achieves substantial performance gains. Cloud-Adapter consistently achieves state-of-the-art performance across various cloud segmentation datasets from multiple satellite sources, sensor series, data processing levels, land cover scenarios, and annotation granularities. We have released the code and model checkpoints at https://xavierjiezou.github.io/Cloud-Adapter/ to support further research.

cs.CV↗

arXiv2024

Detection of the lowest mass ratio contact binary in the universe: TYC 3801-1529-1

This paper presents the first analysis of the contact binary TYC 3801-1529-1. We observed four sets of multiple bands complete light curves and one set of radial velocity curve of the primary component. Based on a simultaneous investigation of our observed and TESS light curves and the radial velocity curve, we found that TYC 3801-1529-1 is an extremely low-mass-ratio, medium contact binary with $q=0.0356$, with the contribution of the third light at a level of about 10\%. Its mass ratio is lower than V1187 Her, making TYC 3801-1529-1 the lowest mass-ratio contact binary ever found in the universe. The light curves observed in 2022 are asymmetric, which is aptly explained by a hot spot on the primary component. A 16-year eclipse timings analysis indicates a secular increase orbital period with a rate of dp/dt$=7.96(\pm0.35)\times10^{-7}$ d yr$^{-1}$. We studied the stability of this target and identified that not only the value of $J_{spin}/J_{orb}$, but also the mass ratio surpass the unstable boundary. Hence, TYC 3801-1529-1 presents a challenge to theoretical research and ought to be considered a progenitor of a contact binary merger.

astro-ph.SR↗

arXiv2024

Exploiting VLM Localizability and Semantics for Open Vocabulary Action Detection

Action detection aims to detect (recognize and localize) human actions spatially and temporally in videos. Existing approaches focus on the closed-set setting where an action detector is trained and tested on videos from a fixed set of action categories. However, this constrained setting is not viable in an open world where test videos inevitably come beyond the trained action categories. In this paper, we address the practical yet challenging Open-Vocabulary Action Detection (OVAD) problem. It aims to detect any action in test videos while training a model on a fixed set of action categories. To achieve such an open-vocabulary capability, we propose a novel method OpenMixer that exploits the inherent semantics and localizability of large vision-language models (VLM) within the family of query-based detection transformers (DETR). Specifically, the OpenMixer is developed by spatial and temporal OpenMixer blocks (S-OMB and T-OMB), and a dynamically fused alignment (DFA) module. The three components collectively enjoy the merits of strong generalization from pre-trained VLMs and end-to-end learning from DETR design. Moreover, we established OVAD benchmarks under various settings, and the experimental results show that the OpenMixer performs the best over baselines for detecting seen and unseen actions. We release the codes, models, and dataset splits at https://github.com/Cogito2012/OpenMixer.

cs.CV↗

arXiv2024

Holographic striped superconductor with ionic lattice

We construct a holographic model to study the striped superconductor on ionic lattices. This model features a phase diagram with three distinct phases, namely the charge density wave (CDW) phase, ordinary superconducting phase (SC) and the striped superconducting phase (SSC). The effect of the ionic lattices on the phase diagram is investigated in detail. First, due to the periodic nature of the background, different types of CDW solutions can be found below the critical temperature. Furthermore, with the increase of the lattice amplitude these solutions are locked in different commensurate states. Second, we find that the critical temperature of CDW phase decreases with the increase of the lattice amplitude, while that of the SC phase increases. Additionally, the background solutions are obtained for different phases, and it is verified that the SSC phase has the lowest free energy among all three phases.

hep-th↗

arXiv2024

A photometric and spectroscopic study of eight semi-detached eclipsing binaries

By cross-matching the eclipsing binary catalog from TESS with that from LAMOST MRS, semi-detached eclipsing binaries with radial velocities coverage spanning more than 0.3 phases were authenticated. The absolute parameters for these systems were determined by simultaneous modeling of light curves and radial velocities using the Wilson-Devinney program. Additionally, the secular orbital variations were further analyzed using O-C curves. Eight semi-detached eclipsing binaries have been identified. Among them, seven feature primary stars situated within the main-sequence band, while their secondaries are all in evolved stages. This suggests that these systems likely originated as detached binaries and have undergone a reversal of the mass ratio. However, TIC 428257299 is an exception where the primary is Roche lobe-filling, and its secondary has experienced mass loss events. Additionally, TIC 8677671 and TIC 318217844 demonstrate secular cyclical changes of orbital periods. Specifically, for TIC 8677671, the cyclical change could result from magnetic activity or a third body which is likely to be compact, with a mass of at least 2.97 M$_{\odot}$.

astro-ph.SR↗

arXiv2024

Diverse Policies Recovering via Pointwise Mutual Information Weighted Imitation Learning

Recovering a spectrum of diverse policies from a set of expert trajectories is an important research topic in imitation learning. After determining a latent style for a trajectory, previous diverse policies recovering methods usually employ a vanilla behavioral cloning learning objective conditioned on the latent style, treating each state-action pair in the trajectory with equal importance. Based on an observation that in many scenarios, behavioral styles are often highly relevant with only a subset of state-action pairs, this paper presents a new principled method in diverse polices recovery. In particular, after inferring or assigning a latent style for a trajectory, we enhance the vanilla behavioral cloning by incorporating a weighting mechanism based on pointwise mutual information. This additional weighting reflects the significance of each state-action pair's contribution to learning the style, thus allowing our method to focus on state-action pairs most representative of that style. We provide theoretical justifications for our new objective, and extensive empirical evaluations confirm the effectiveness of our method in recovering diverse policies from expert data.

cs.LG↗

arXiv2024

BitDelta: Your Fine-Tune May Only Be Worth One Bit

Large Language Models (LLMs) are typically trained in two phases: pre-training on large internet-scale datasets, and fine-tuning for downstream tasks. Given the higher computational demand of pre-training, it's intuitive to assume that fine-tuning adds less new information to the model, and is thus more compressible. We explore this assumption by decomposing the weights of fine-tuned models into their pre-trained components and an additional delta. We introduce a simple method, BitDelta, which successfully quantizes this delta down to 1 bit without compromising performance. This interesting finding not only highlights the potential redundancy of information added during fine-tuning, but also has significant implications for the multi-tenant serving and multi-tenant storage of fine-tuned models. By enabling the use of a single high-precision base model accompanied by multiple 1-bit deltas, BitDelta dramatically reduces GPU memory requirements by more than 10x, which can also be translated to enhanced generation latency in multi-tenant settings. We validate BitDelta through experiments across Llama-2 and Mistral model families, and on models up to 70B parameters, showcasing minimal performance degradation over all tested settings.

cs.LG↗

arXiv2024

Evaluating Copyright Takedown Methods for Language Models

Language models (LMs) derive their capabilities from extensive training on diverse data, including potentially copyrighted material. These models can memorize and generate content similar to their training data, posing potential concerns. Therefore, model creators are motivated to develop mitigation methods that prevent generating protected content. We term this procedure as copyright takedowns for LMs, noting the conceptual similarity to (but legal distinction from) the DMCA takedown This paper introduces the first evaluation of the feasibility and side effects of copyright takedowns for LMs. We propose CoTaEval, an evaluation framework to assess the effectiveness of copyright takedown methods, the impact on the model's ability to retain uncopyrightable factual knowledge from the training data whose recitation is embargoed, and how well the model maintains its general utility and efficiency. We examine several strategies, including adding system prompts, decoding-time filtering interventions, and unlearning approaches. Our findings indicate that no tested method excels across all metrics, showing significant room for research in this unique problem setting and indicating potential unresolved challenges for live policy proposals.

cs.CL↗

arXiv2024

Adapter-dependent Adapter Methylation Assay

Sensitive and reliable methylation assay is important for oncogentic studies and clinical applications. Here, a new methylation assay was developed by the use of adapter-dependent adapter in library preparation. This new assay avoids the use of bisulfite and provides a simple and highly sensitive scanning of methylation spectra of circulating free DNA and genomic DNA.

q-bio.BM↗

arXiv2024

A Novel Framework of Horizontal-Vertical Hybrid Federated Learning for EdgeIoT

This letter puts forth a new hybrid horizontal-vertical federated learning (HoVeFL) for mobile edge computing-enabled Internet of Things (EdgeIoT). In this framework, certain EdgeIoT devices train local models using the same data samples but analyze disparate data features, while the others focus on the same features using non-independent and identically distributed (non-IID) data samples. Thus, even though the data features are consistent, the data samples vary across devices. The proposed HoVeFL formulates the training of local and global models to minimize the global loss function. Performance evaluations on CIFAR-10 and SVHN datasets reveal that the testing loss of HoVeFL with 12 horizontal FL devices and six vertical FL devices is 5.5% and 25.2% higher, respectively, compared to a setup with six horizontal FL devices and 12 vertical FL devices.

cs.DC↗