arXiv subjects

Kun Zhao

Publications and source records attributed to Kun Zhao.

At least 55 records · Page 3Linked to original sources

arXiv2024

MeloTrans: A Text to Symbolic Music Generation Model Following Human Composition Habit

At present, neural network models show powerful sequence prediction ability and are used in many automatic composition models. In comparison, the way humans compose music is very different from it. Composers usually start by creating musical motifs and then develop them into music through a series of rules. This process ensures that the music has a specific structure and changing pattern. However, it is difficult for neural network models to learn these composition rules from training data, which results in a lack of musicality and diversity in the generated music. This paper posits that integrating the learning capabilities of neural networks with human-derived knowledge may lead to better results. To archive this, we develop the POP909$\_$M dataset, the first to include labels for musical motifs and their variants, providing a basis for mimicking human compositional habits. Building on this, we propose MeloTrans, a text-to-music composition model that employs principles of motif development rules. Our experiments demonstrate that MeloTrans excels beyond existing music generation models and even surpasses Large Language Models (LLMs) like ChatGPT-4. This highlights the importance of merging human insights with neural network capabilities to achieve superior symbolic music generation.

cs.SD↗

arXiv2024

CASPFormer: Trajectory Prediction from BEV Images with Deformable Attention

Motion prediction is an important aspect for Autonomous Driving (AD) and Advance Driver Assistance Systems (ADAS). Current state-of-the-art motion prediction methods rely on High Definition (HD) maps for capturing the surrounding context of the ego vehicle. Such systems lack scalability in real-world deployment as HD maps are expensive to produce and update in real-time. To overcome this issue, we propose Context Aware Scene Prediction Transformer (CASPFormer), which can perform multi-modal motion prediction from rasterized Bird-Eye-View (BEV) images. Our system can be integrated with any upstream perception module that is capable of generating BEV images. Moreover, CASPFormer directly decodes vectorized trajectories without any postprocessing. Trajectories are decoded recurrently using deformable attention, as it is computationally efficient and provides the network with the ability to focus its attention on the important spatial locations of the BEV images. In addition, we also address the issue of mode collapse for generating multiple scene-consistent trajectories by incorporating learnable mode queries. We evaluate our model on the nuScenes dataset and show that it reaches state-of-the-art across multiple metrics

cs.LG↗

arXiv2024

Autocorrelation Measurement of Attosecond Pulses Based on Two-Photon Double Ionization

Autocorrelation measurement is theoretically demonstrated to characterize attosecond pulses by studying the two-photon double ionization (TPDI) process. An interferometric autocorrelation curve is presented in the change of TPDI probability with the time delay between two identical attosecond pulses, and its full width at half maximum (FWHM) $τ_{e}$ has a relationship $τ_{e}=1.77τ+15$ with the FWHM $τ$ of the attosecond pulse. The curve is also decoded to obtain the center frequency and FWHM of the attosecond pulse by fitting. In addition, the required peak intensity of the attosecond pulse is estimated to be on the order of $10^{16}\,\rm{Wcm^{-2}}$ in autocorrelation experiments. The findings pave the way for autocorrelation measurement of intense isolated attosecond pulses.

physics.atom-ph↗

arXiv2024

Predicting T-Cell Receptor Specificity

Researching the specificity of TCR contributes to the development of immunotherapy and provides new opportunities and strategies for personalized cancer immunotherapy. Therefore, we established a TCR generative specificity detection framework consisting of an antigen selector and a TCR classifier based on the Random Forest algorithm, aiming to efficiently screen out TCRs and target antigens and achieve TCR specificity prediction. Furthermore, we used the k-fold validation method to compare the performance of our model with ordinary deep learning methods. The result proves that adding a classifier to the model based on the random forest algorithm is very effective, and our model generally outperforms ordinary deep learning methods. Moreover, we put forward feasible optimization suggestions for the shortcomings and challenges of our model found during model implementation.

q-bio.QM↗

arXiv2024

Panoptic Segmentation of Mammograms with Text-To-Image Diffusion Model

Mammography is crucial for breast cancer surveillance and early diagnosis. However, analyzing mammography images is a demanding task for radiologists, who often review hundreds of mammograms daily, leading to overdiagnosis and overtreatment. Computer-Aided Diagnosis (CAD) systems have been developed to assist in this process, but their capabilities, particularly in lesion segmentation, remained limited. With the contemporary advances in deep learning their performance may be improved. Recently, vision-language diffusion models emerged, demonstrating outstanding performance in image generation and transferability to various downstream tasks. We aim to harness their capabilities for breast lesion segmentation in a panoptic setting, which encompasses both semantic and instance-level predictions. Specifically, we propose leveraging pretrained features from a Stable Diffusion model as inputs to a state-of-the-art panoptic segmentation architecture, resulting in accurate delineation of individual breast lesions. To bridge the gap between natural and medical imaging domains, we incorporated a mammography-specific MAM-E diffusion model and BiomedCLIP image and text encoders into this framework. We evaluated our approach on two recently published mammography datasets, CDD-CESM and VinDr-Mammo. For the instance segmentation task, we noted 40.25 AP0.1 and 46.82 AP0.05, as well as 25.44 PQ0.1 and 26.92 PQ0.05. For the semantic segmentation task, we achieved Dice scores of 38.86 and 40.92, respectively.

cs.CV↗

arXiv2024

BioMNER: A Dataset for Biomedical Method Entity Recognition

Named entity recognition (NER) stands as a fundamental and pivotal task within the realm of Natural Language Processing. Particularly within the domain of Biomedical Method NER, this task presents notable challenges, stemming from the continual influx of domain-specific terminologies in scholarly literature. Current research in Biomedical Method (BioMethod) NER suffers from a scarcity of resources, primarily attributed to the intricate nature of methodological concepts, which necessitate a profound understanding for precise delineation. In this study, we propose a novel dataset for biomedical method entity recognition, employing an automated BioMethod entity recognition and information retrieval system to assist human annotation. Furthermore, we comprehensively explore a range of conventional and contemporary open-domain NER methodologies, including the utilization of cutting-edge large-scale language models (LLMs) customised to our dataset. Our empirical findings reveal that the large parameter counts of language models surprisingly inhibit the effective assimilation of entity extraction patterns pertaining to biomedical methods. Remarkably, the approach, leveraging the modestly sized ALBERT model (only 11MB), in conjunction with conditional random fields (CRF), achieves state-of-the-art (SOTA) performance.

cs.CL↗

arXiv2024

SLIDE: A Framework Integrating Small and Large Language Models for Open-Domain Dialogues Evaluation

The long-standing one-to-many problem of gold standard responses in open-domain dialogue systems presents challenges for automatic evaluation metrics. Though prior works have demonstrated some success by applying powerful Large Language Models (LLMs), existing approaches still struggle with the one-to-many problem, and exhibit subpar performance in domain-specific scenarios. We assume the commonsense reasoning biases within LLMs may hinder their performance in domainspecific evaluations. To address both issues, we propose a novel framework SLIDE (Small and Large Integrated for Dialogue Evaluation), that leverages both a small, specialised model (SLM), and LLMs for the evaluation of open domain dialogues. Our approach introduces several techniques: (1) Contrastive learning to differentiate between robust and non-robust response embeddings; (2) A novel metric for semantic sensitivity that combines embedding cosine distances with similarity learned through neural networks, and (3) a strategy for incorporating the evaluation results from both the SLM and LLMs. Our empirical results demonstrate that our approach achieves state-of-the-art performance in both the classification and evaluation tasks, and additionally the SLIDE evaluator exhibits better correlation with human judgements. Our code is available at https:// github.com/hegehongcha/SLIDE-ACL2024.

cs.CL↗

arXiv2024

Interpretable Spatio-Temporal Embedding for Brain Structural-Effective Network with Ordinary Differential Equation

The MRI-derived brain network serves as a pivotal instrument in elucidating both the structural and functional aspects of the brain, encompassing the ramifications of diseases and developmental processes. However, prevailing methodologies, often focusing on synchronous BOLD signals from functional MRI (fMRI), may not capture directional influences among brain regions and rarely tackle temporal functional dynamics. In this study, we first construct the brain-effective network via the dynamic causal model. Subsequently, we introduce an interpretable graph learning framework termed Spatio-Temporal Embedding ODE (STE-ODE). This framework incorporates specifically designed directed node embedding layers, aiming at capturing the dynamic interplay between structural and effective networks via an ordinary differential equation (ODE) model, which characterizes spatial-temporal brain dynamics. Our framework is validated on several clinical phenotype prediction tasks using two independent publicly available datasets (HCP and OASIS). The experimental results clearly demonstrate the advantages of our model compared to several state-of-the-art methods.

cs.LG↗

arXiv2024

Effective Distillation of Table-based Reasoning Ability from LLMs

Large Language Models (LLMs) have demonstrated remarkable performance across a wide range of natural language processing tasks. However, their enormous parameter size and extremely high requirements for compute power pose challenges for their practical deployment. Recent research has revealed that specific capabilities of LLMs, such as numerical reasoning, can be transferred to smaller models through distillation. Some studies explore the potential of leveraging LLMs to perform table-based reasoning. However, there has been no prior work focusing on table reasoning skills in smaller models specifically tailored for scientific table-to-text generation tasks. In this paper, we propose a novel table-based reasoning distillation approach, with the aim of distilling LLMs into tailored smaller models. Our experimental results have shown that a 220 million parameter model (Flan-T5-base) fine-tuned using distilled data, not only achieves a significant improvement compared to traditionally fine-tuned baselines, but also surpasses specific LLMs on a scientific table-to-text generation dataset. Our code is available at https://github.com/Bernard-Yang/DistillTableCoT.

cs.CL↗

arXiv2024

Spatially Selective Reconfigurable Intelligent Surfaces Through Element Permutation

A standard reconfigurable intelligent surface (RIS) can be configured to reflect signals from an arbitrary impinging direction to an arbitrary outgoing direction. However, if a signal impinges from any other direction, said signal is reflected, with full beamforming gain, to a specific direction, which is easily determined. The goal of this paper is to propose a RIS which \emph{only} reflects signals from the configured impinging direction. This can be accomplished by a RIS architecture that permutes the antenna elements in the sense that a signal is re-radiated from a different antenna than the one receiving the signal. We analytically prove this fact, and also discuss several variants and hardware implementations.

eess.SP↗

arXiv2024

Constrained Multiview Representation for Self-supervised Contrastive Learning

Representation learning constitutes a pivotal cornerstone in contemporary deep learning paradigms, offering a conduit to elucidate distinctive features within the latent space and interpret the deep models. Nevertheless, the inherent complexity of anatomical patterns and the random nature of lesion distribution in medical image segmentation pose significant challenges to the disentanglement of representations and the understanding of salient features. Methods guided by the maximization of mutual information, particularly within the framework of contrastive learning, have demonstrated remarkable success and superiority in decoupling densely intertwined representations. However, the effectiveness of contrastive learning highly depends on the quality of the positive and negative sample pairs, i.e. the unselected average mutual information among multi-views would obstruct the learning strategy so the selection of the views is vital. In this work, we introduce a novel approach predicated on representation distance-based mutual information (MI) maximization for measuring the significance of different views, aiming at conducting more efficient contrastive learning and representation disentanglement. Additionally, we introduce an MI re-ranking strategy for representation selection, benefiting both the continuous MI estimating and representation significance distance measuring. Specifically, we harness multi-view representations extracted from the frequency domain, re-evaluating their significance based on mutual information across varying frequencies, thereby facilitating a multifaceted contrastive learning approach to bolster semantic comprehension. The statistical results under the five metrics demonstrate that our proposed framework proficiently constrains the MI maximization-driven representation selection and steers the multi-view contrastive learning process.

cs.CV↗

arXiv2024

Many-body hybrid Excitons in Organic-Inorganic van der Waals Heterostructures

The coherent many-body interaction at the organic-inorganic interface can give rise to intriguing hybrid excitons that combine the advantages of the Wannier-Mott and Frenkel excitons simultaneously. Unlike the 2D inorganic heterostructures that suffer from moment mismatch, the hybrid excitons formed at the organic-inorganic interface have a momentum-direct nature, which have yet to be explored. Here, we report hybrid excitons at the copper phthalocyanine/molybdenum diselenide (CuPc/MoSe2) interface with strong molecular orientation dependence using low-temperature photoluminescence spectroscopy. The new emission peaks observed in the CuPc/MoSe2 heterostructure indicate the formation of interfacial hybrid excitons. The density functional theory (DFT) calculation confirms the strong hybridization between the lowest unoccupied molecular orbital (LUMO) of CuPc and the conduction band minimum (CBM) of MoSe2, suggesting that the hybrid excitons consist of electrons extended in both layers and holes confined in individual layers. The temperature-dependent measurements show that the hybrid excitons can gain the signatures of the Frenkel excitons of CuPc and the Wannier-Mott excitons of MoSe2 simultaneously. The out-of-plane molecular orientation is used to tailor the interfacial hybrid exciton states. Our results reveal the hybrid excitons at the CuPc/MoSe2 interface with tunability by molecular orientation, which suggests that the emerging organic-inorganic heterostructure can be a promising platform for many-body exciton physics.

physics.optics↗

arXiv2023

Delivery of nanosecond laser pulses by multimode anti-resonant hollow core fiber at 1 um wavelength

In this paper we explore the application of low-loss multimode anti-resonant hollow-core fiber (MM-AR-HCF) in the delivery of nanosecond laser pulses at 1 um wavelength. MM-AR-HCF of large core offers a rich content of low-loss higher-order modes which plays a key role in the efficient coupling and transmission of high-power laser of degraded beam quality. In the experiment, laser pulses of an average pulse energy of 21.8 mJ with 14.6 ns pulse width (corresponding a peak power of 1.49 MW) are transmitted through MM-AR-HCF of 9.8 m length without damaging. Up to 94 % coupling efficiency is achieved where the incident laser beam suffers a degraded beam quality with and of 2.18 and 1.99 respectively. Laser-induced damage threshold (LIDT) of MM-AR-HCF measures 22.6 mJ for 94 % coupling efficiency, which is 7 times higher than that for multimode silica optical fiber with a core diameter of 200 um.

physics.optics↗

arXiv2023

CASPNet++: Joint Multi-Agent Motion Prediction

The prediction of road users' future motion is a critical task in supporting advanced driver-assistance systems (ADAS). It plays an even more crucial role for autonomous driving (AD) in enabling the planning and execution of safe driving maneuvers. Based on our previous work, Context-Aware Scene Prediction Network (CASPNet), an improved system, CASPNet++, is proposed. In this work, we focus on further enhancing the interaction modeling and scene understanding to support the joint prediction of all road users in a scene using spatiotemporal grids to model future occupancy. Moreover, an instance-based output head is introduced to provide multi-modal trajectories for agents of interest. In extensive quantitative and qualitative analysis, we demonstrate the scalability of CASPNet++ in utilizing and fusing diverse environmental input sources such as HD maps, Radar detection, and Lidar segmentation. Tested on the urban-focused prediction dataset nuScenes, CASPNet++ reaches state-of-the-art performance. The model has been deployed in a testing vehicle, running in real-time with moderate computational resources.

cs.CV↗

arXiv2023

Evaluating Open-Domain Dialogues in Latent Space with Next Sentence Prediction and Mutual Information

The long-standing one-to-many issue of the open-domain dialogues poses significant challenges for automatic evaluation methods, i.e., there may be multiple suitable responses which differ in semantics for a given conversational context. To tackle this challenge, we propose a novel learning-based automatic evaluation metric (CMN), which can robustly evaluate open-domain dialogues by augmenting Conditional Variational Autoencoders (CVAEs) with a Next Sentence Prediction (NSP) objective and employing Mutual Information (MI) to model the semantic similarity of text in the latent space. Experimental results on two open-domain dialogue datasets demonstrate the superiority of our method compared with a wide range of baselines, especially in handling responses which are distant to the golden reference responses in semantics.

cs.CL↗

arXiv2023

Global stability of a logarithmically sensitive chemotaxis model under time-dependent boundary conditions

This paper studies the dynamical behavior of classical solutions to a hyperbolic system of balance laws, derived from a chemotaxis model with logarithmic sensitivity, subject to time-dependent boundary conditions. It is shown that under suitable assumptions on the boundary data, solutions starting in $H^2$-space exist globally in time and the differences between the solutions and their corresponding boundary data converge to zero, as time goes to infinity. There is no smallness restriction on the magnitude of initial perturbations. Moreover, numerical simulations show that the assumptions on the boundary data are necessary for the above mentioned results.

math.AP↗

arXiv2023

Semantic Interleaving Global Channel Attention for Multilabel Remote Sensing Image Classification

Multi-Label Remote Sensing Image Classification (MLRSIC) has received increasing research interest. Taking the cooccurrence relationship of multiple labels as additional information helps to improve the performance of this task. Current methods focus on using it to constrain the final feature output of a Convolutional Neural Network (CNN). On the one hand, these methods do not make full use of label correlation to form feature representation. On the other hand, they increase the label noise sensitivity of the system, resulting in poor robustness. In this paper, a novel method called Semantic Interleaving Global Channel Attention (SIGNA) is proposed for MLRSIC. First, the label co-occurrence graph is obtained according to the statistical information of the data set. The label co-occurrence graph is used as the input of the Graph Neural Network (GNN) to generate optimal feature representations. Then, the semantic features and visual features are interleaved, to guide the feature expression of the image from the original feature space to the semantic feature space with embedded label relations. SIGNA triggers global attention of feature maps channels in a new semantic feature space to extract more important visual features. Multihead SIGNA based feature adaptive weighting networks are proposed to act on any layer of CNN in a plug-and-play manner. For remote sensing images, better classification performance can be achieved by inserting CNN into the shallow layer. We conduct extensive experimental comparisons on three data sets: UCM data set, AID data set, and DFC15 data set. Experimental results demonstrate that the proposed SIGNA achieves superior classification performance compared to state-of-the-art (SOTA) methods. It is worth mentioning that the codes of this paper will be open to the community for reproducibility research. Our codes are available at https://github.com/kyle-one/SIGNA.

cs.CV↗

arXiv2023

Credible Remote Sensing Scene Classification Using Evidential Fusion on Aerial-Ground Dual-view Images

Due to their ability to offer more comprehensive information than data from a single view, multi-view (multi-source, multi-modal, multi-perspective, etc.) data are being used more frequently in remote sensing tasks. However, as the number of views grows, the issue of data quality becomes more apparent, limiting the potential benefits of multi-view data. Although recent deep neural network (DNN) based models can learn the weight of data adaptively, a lack of research on explicitly quantifying the data quality of each view when fusing them renders these models inexplicable, performing unsatisfactorily and inflexible in downstream remote sensing tasks. To fill this gap, in this paper, evidential deep learning is introduced to the task of aerial-ground dual-view remote sensing scene classification to model the credibility of each view. Specifically, the theory of evidence is used to calculate an uncertainty value which describes the decision-making risk of each view. Based on this uncertainty, a novel decision-level fusion strategy is proposed to ensure that the view with lower risk obtains more weight, making the classification more credible. On two well-known, publicly available datasets of aerial-ground dual-view remote sensing images, the proposed approach achieves state-of-the-art results, demonstrating its effectiveness. The code and datasets of this article are available at the following address: https://github.com/gaopiaoliang/Evidential.

cs.CV↗