arXiv subjects

Chao Zhang

Publications and source records attributed to Chao Zhang.

At least 541 records · Page 30Linked to original sources

arXiv2024

MSSC-BiMamba: Multimodal Sleep Stage Classification and Early Diagnosis of Sleep Disorders with Bidirectional Mamba

Monitoring sleep states is essential for evaluating sleep quality and diagnosing sleep disorders. Traditional manual staging is time-consuming and prone to subjective bias, often resulting in inconsistent outcomes. Here, we developed an automated model for sleep staging and disorder classification to enhance diagnostic accuracy and efficiency. Considering the characteristics of polysomnography (PSG) multi-lead sleep monitoring, we designed a multimodal sleep state classification model, MSSC-BiMamba, that combines an Efficient Channel Attention (ECA) mechanism with a Bidirectional State Space Model (BSSM). The ECA module allows for weighting data from different sensor channels, thereby amplifying the influence of diverse sensor inputs. Additionally, the implementation of bidirectional Mamba (BiMamba) enables the model to effectively capture the multidimensional features and long-range dependencies of PSG data. The developed model demonstrated impressive performance on sleep stage classification tasks on both the ISRUC-S3 and ISRUC-S1 datasets, respectively containing data with healthy and unhealthy sleep patterns. Also, the model exhibited a high accuracy for sleep health prediction when evaluated on a combined dataset consisting of ISRUC and Sleep-EDF. Our model, which can effectively handle diverse sleep conditions, is the first to apply BiMamba to sleep staging with multimodal PSG data, showing substantial gains in computational and memory efficiency over traditional Transformer-style models. This method enhances sleep health management by making monitoring more accessible and extending advanced healthcare through innovative technology.

cs.AI↗

arXiv2024

On the weak Harnack inequalities for nonlocal double phase problems

This paper is devoted to studying the weak Harnack inequalities for nonlocal double phase functionals by using expansion of positivity, whose prototype is $$ \iint_{\mathbb{R}^n\times\mathbb{R}^n} \left(\frac{|u(x)-u(y)|^p}{|x-y|^{n+sp}}+a(x,y)\frac{|u(x)-u(y)|^q}{|x-y|^{n+tq}}\right) \,dxdy $$ with $a\ge0$ and $0<s\le t<1<p\le q$. The core of our approach is to establish several measure theoretical estimates based on the nonlocal Caccioppoli-type inequality, where the challenges consist in controlling subtle interaction between the pointwise behaviour of modulating coefficient and the growth exponents. Meanwhile, a quantitative boundedness result on the minimizer of such functionals is also discussed.

math.AP↗

arXiv2024

BBox-Adapter: Lightweight Adapting for Black-Box Large Language Models

Adapting state-of-the-art Large Language Models (LLMs) like GPT-4 and Gemini for specific tasks is challenging. Due to the opacity in their parameters, embeddings, and even output probabilities, existing fine-tuning adaptation methods are inapplicable. Consequently, adapting these black-box LLMs is only possible through their API services, raising concerns about transparency, privacy, and cost. To address these challenges, we introduce BBox-Adapter, a novel lightweight adapter for black-box LLMs. BBox-Adapter distinguishes target and source domain data by treating target data as positive and source data as negative. It employs a ranking-based Noise Contrastive Estimation (NCE) loss to promote the likelihood of target domain data while penalizing that of the source domain. Furthermore, it features an online adaptation mechanism, which incorporates real-time positive data sampling from ground-truth, human, or AI feedback, coupled with negative data from previous adaptations. Extensive experiments demonstrate BBox-Adapter's effectiveness and cost efficiency. It improves model performance by up to 6.77% across diverse tasks and domains, while reducing training and inference costs by 31.30x and 1.84x, respectively.

cs.CL↗

arXiv2024

Dipolar bosons in a twisted bilayer geometry

In recent years, twisted bilayer systems such as bilayer graphene have attracted a great deal of attention as the twist angle introduces a degree of freedom which can be used to non-trivially modify system properties. This idea has been picked up in the cold atom community, first with a theoretical proposal to simulate twisted bilayers in state-dependent optical lattices, and, more recently, with an experimental realization of twisted bilayers with bosonic atoms in two different spin states. In this manuscript, we theoretically investigate dipolar bosons in a twisted bilayer geometry. The interplay between dipolar interaction and the twist between the layers results in the emergence of quantum states not observed in the absence of twist. We study how system properties vary as we change the twist angle at fixed distance between the layers and fixed dipolar interaction. We find that at a twist angle $θ=0.1^{\circ}$, the observed quantum phases are consistent with those seen in the absence of twist angle, i.e. paired superfluid, paired supersolid, and paired solid phases. However, a slight increase in the twist angle to $θ=0.2^{\circ}$ disrupts these paired phases in favor of a phase separation between checkerboard solid and superfluid regions. Notably, at a twist angle of $θ=5.21^{\circ}$, the local occupation number follows the moiré pattern of the underlying moiré bilayers so that a periodic structure of insulating islands is formed. These insulating islands are surrounded by a superfluid.

cond-mat.quant-gas↗

arXiv2024

Incremental Pseudo-Labeling for Black-Box Unsupervised Domain Adaptation

Black-Box unsupervised domain adaptation (BBUDA) learns knowledge only with the prediction of target data from the source model without access to the source data and source model, which attempts to alleviate concerns about the privacy and security of data. However, incorrect pseudo-labels are prevalent in the prediction generated by the source model due to the cross-domain discrepancy, which may substantially degrade the performance of the target model. To address this problem, we propose a novel approach that incrementally selects high-confidence pseudo-labels to improve the generalization ability of the target model. Specifically, we first generate pseudo-labels using a source model and train a crude target model by a vanilla BBUDA method. Second, we iteratively select high-confidence data from the low-confidence data pool by thresholding the softmax probabilities, prototype labels, and intra-class similarity. Then, we iteratively train a stronger target network based on the crude target model to correct the wrongly labeled samples to improve the accuracy of the pseudo-label. Experimental results demonstrate that the proposed method achieves state-of-the-art black-box unsupervised domain adaptation performance on three benchmark datasets.

cs.CV↗

arXiv2024

Self-Taught Recognizer: Toward Unsupervised Adaptation for Speech Foundation Models

We propose an unsupervised adaptation framework, Self-TAught Recognizer (STAR), which leverages unlabeled data to enhance the robustness of automatic speech recognition (ASR) systems in diverse target domains, such as noise and accents. STAR is developed for prevalent speech foundation models based on Transformer-related architecture with auto-regressive decoding (e.g., Whisper, Canary). Specifically, we propose a novel indicator that empirically integrates step-wise information during decoding to assess the token-level quality of pseudo labels without ground truth, thereby guiding model updates for effective unsupervised adaptation. Experimental results show that STAR achieves an average of 13.5% relative reduction in word error rate across 14 target domains, and it sometimes even approaches the upper-bound performance of supervised adaptation. Surprisingly, we also observe that STAR prevents the adapted model from the common catastrophic forgetting problem without recalling source-domain data. Furthermore, STAR exhibits high data efficiency that only requires less than one-hour unlabeled data, and seamless generality to alternative large speech models and speech translation tasks. Our code aims to open source to the research communities.

cs.CL↗

arXiv2024

CrossCheckGPT: Universal Hallucination Ranking for Multimodal Foundation Models

Multimodal foundation models are prone to hallucination, generating outputs that either contradict the input or are not grounded by factual information. Given the diversity in architectures, training data and instruction tuning techniques, there can be large variations in systems' susceptibility to hallucinations. To assess system hallucination robustness, hallucination ranking approaches have been developed for specific tasks such as image captioning, question answering, summarization, or biography generation. However, these approaches typically compare model outputs to gold-standard references or labels, limiting hallucination benchmarking for new domains. This work proposes "CrossCheckGPT", a reference-free universal hallucination ranking for multimodal foundation models. The core idea of CrossCheckGPT is that the same hallucinated content is unlikely to be generated by different independent systems, hence cross-system consistency can provide meaningful and accurate hallucination assessment scores. CrossCheckGPT can be applied to any model or task, provided that the information consistency between outputs can be measured through an appropriate distance metric. Focusing on multimodal large language models that generate text, we explore two information consistency measures: CrossCheck-explicit and CrossCheck-implicit. We showcase the applicability of our method for hallucination ranking across various modalities, namely the text, image, and audio-visual domains. Further, we propose the first audio-visual hallucination benchmark, "AVHalluBench", and illustrate the effectiveness of CrossCheckGPT, achieving correlations of 98% and 89% with human judgements on MHaluBench and AVHalluBench, respectively.

cs.CL↗

arXiv2024

Double phase anisotropic variational problems involving critical growth

In this paper, we investigate some existence results for double phase anisotropic variational problems involving critical growth. We first establish a Lions type concentration-compactness principle and its variant at infinity for the solution space, which are our independent interests. By employing these results, we obtain a nontrivial nonnegative solution to problems of generalized concave-convex type. We also obtain infinitely many solutions when the nonlinear term is symmetric. Our results are new even for the $p(\cdot)$-Laplace equations.

math.AP↗

arXiv2024

Hunyuan-DiT: A Powerful Multi-Resolution Diffusion Transformer with Fine-Grained Chinese Understanding

We present Hunyuan-DiT, a text-to-image diffusion transformer with fine-grained understanding of both English and Chinese. To construct Hunyuan-DiT, we carefully design the transformer structure, text encoder, and positional encoding. We also build from scratch a whole data pipeline to update and evaluate data for iterative model optimization. For fine-grained language understanding, we train a Multimodal Large Language Model to refine the captions of the images. Finally, Hunyuan-DiT can perform multi-turn multimodal dialogue with users, generating and refining images according to the context. Through our holistic human evaluation protocol with more than 50 professional human evaluators, Hunyuan-DiT sets a new state-of-the-art in Chinese-to-image generation compared with other open-source models. Code and pretrained models are publicly available at github.com/Tencent/HunyuanDiT

cs.CV↗

arXiv2024

Goal-oriented compression for $L_p$-norm-type goal functions: Application to power consumption scheduling

Conventional data compression schemes aim at implementing a trade-off between the rate required to represent the compressed data and the resulting distortion between the original and reconstructed data. However, in more and more applications, what is desired is not reconstruction accuracy but the quality of the realization of a certain task by the receiver. In this paper, the receiver task is modeled by an optimization problem whose parameters have to be compressed by the transmitter. Motivated by applications such as the smart grid, this paper focuses on a goal function which is of $L_p$-norm-type. The aim is to design the precoding, quantization, and decoding stages such that the maximum of the goal function obtained with the compressed version of the parameters is as close as possible to the maximum obtained without compression. The numerical analysis, based on real smart grid signals, clearly shows the benefits of the proposed approach compared to the conventional distortion-based compression paradigm.

eess.SP↗

arXiv2024

On monomial algebras with representation-finite enveloping algebras

The present paper mainly considers the representation type of the enveloping algebra of monomial algebra. Let $A$ be a monomial algebra and $A^e= A\otimes_{\mathrm{l}\!\mathrm{k}} A^{\mathrm{op}}$ its enveloping algebra. It is shown that $A^e$ is representation-finite if and only if $A \cong \pmb{A}_n/\mathrm{rad}^2 \pmb{A}_n$, where $\pmb{A}_n$ is the path algebra $\mathrm{l}\!\mathrm{k}\mathcal{Q}$ with $\mathcal{Q} = 1 \longrightarrow 2 \longrightarrow \cdots \longrightarrow n$. Moreover, we show that the number of all isoclasses of indecomposable $(\pmb{A}_n/ \mathrm{rad}^2\pmb{A}_n)^e$-modules is $\frac{4}{3}n^3 + n^2-\frac{7}{3}n+1$, and classify all indecomposable modules over $(\pmb{A}_n/ \mathrm{rad}^2\pmb{A}_n)^e$. Finally, the Clebsch-Gordon problem over $(\pmb{A}_n/ \mathrm{rad}^2\pmb{A}_n)^e$ is studied.

math.RT↗

arXiv2024

Frequency-Guided Multi-Level Human Action Anomaly Detection with Normalizing Flows

We introduce the task of human action anomaly detection (HAAD), which aims to identify anomalous motions in an unsupervised manner given only the pre-determined normal category of training action samples. Compared to prior human-related anomaly detection tasks which primarily focus on unusual events from videos, HAAD involves the learning of specific action labels to recognize semantically anomalous human behaviors. To address this task, we propose a normalizing flow (NF)-based detection framework where the sample likelihood is effectively leveraged to indicate anomalies. As action anomalies often occur in some specific body parts, in addition to the full-body action feature learning, we incorporate extra encoding streams into our framework for a finer modeling of body subsets. Our framework is thus multi-level to jointly discover global and local motion anomalies. Furthermore, to show awareness of the potentially jittery data during recording, we resort to discrete cosine transformation by converting the action samples from the temporal to the frequency domain to mitigate the issue of data instability. Extensive experimental results on two human action datasets demonstrate that our method outperforms the baselines formed by adapting state-of-the-art human activity AD approaches to our task of HAAD.

cs.CV↗

arXiv2024

Genuine multipartite entanglement detection with imperfect measurements: concept and experiment

Standard procedures for entanglement detection assume that experimenters can exactly implement specific quantum measurements. Here, we depart from such idealizations and investigate, in both theory and experiment, the detection of genuine multipartite entanglement when measurements are subject to small imperfections. For arbitrary qubits number $n$, we construct multipartite entanglement witnesses where the detrimental influence of the imperfection is independent of $n$. In a tabletop four-partite photonic experiment we demonstrate first how a small amount of alignment error can undermine the conclusions drawn from standard entanglement witnesses, and then perform the correction analysis. Furthermore, since we consider quantum devices that are trusted but not perfectly controlled, we showcase advantages in terms of noise resilience as compared to device-independent models.

quant-ph↗

arXiv2024

HTAP Databases: A Survey

Since Gartner coined the term, Hybrid Transactional and Analytical Processing (HTAP), numerous HTAP databases have been proposed to combine transactions with analytics in order to enable real-time data analytics for various data-intensive applications. HTAP databases typically process the mixed workloads of transactions and analytical queries in a unified system by leveraging both a row store and a column store. As there are different storage architectures and processing techniques to satisfy various requirements of diverse applications, it is critical to summarize the pros and cons of these key techniques. This paper offers a comprehensive survey of HTAP databases. We mainly classify state-of-the-art HTAP databases according to four storage architectures: (a) Primary Row Store and In-Memory Column Store; (b) Distributed Row Store and Column Store Replica; (c) Primary Row Store and Distributed In-Memory Column Store; and (d) Primary Column Store and Delta Row Store. We then review the key techniques in HTAP databases, including hybrid workload processing, data organization, data synchronization, query optimization, and resource scheduling. We also discuss existing HTAP benchmarks. Finally, we provide the research challenges and opportunities for HTAP techniques.

cs.DB↗

arXiv2024

Emerging Advancements in 6G NTN Radio Access Technologies: An Overview

The efforts on the development, standardization and improvements to communication systems towards 5G Advanced and 6G are on track to provide benefits such as an unprecedented level of connectivity and performance, enabling a diverse range of vertical services. The full integration of non-terrestrial components into 6G plays a pivotal role in realizing this paradigm shift towards ubiquitous communication and global coverage. However, this integration into 6G brings forth a set of its own challenges, particularly in Radio Access Technologies (RATs). To this end, this paper comprehensively discusses those challenges at different levels of RATs and proposes the corresponding potential emerging advancements in the realm of 6G NTN. In particular, the focus is on advancing the prospective aspects of Radio Resource Management (RRM), spectral coexistence in terrestrial and non-terrestrial components and flexible waveform design solutions to combat the impediments. This discussion with a specific focus on emerging advancements in 6G NTN RATs is critical for shaping the next generation networks and potentially relevant in contributing the part in standardization in forthcoming releases

eess.SP↗

arXiv2024

Loss of coherence and change in emission physics for radio emission from very inclined cosmic-ray air showers

Next-generation radio experiments such as the Radio Detector of the upgraded Pierre Auger Observatory and the planned GRAND and BEACON arrays target the detection of ultra-high-energy particle air showers arriving at low elevation angles. These inclined cosmic-ray air showers develop higher in the atmosphere than vertical ones, enhancing magnetic deflections of electrons and positrons inside the cascade. We evidence two novel features in their radio emission: a new polarization pattern, consistent with a geo-synchrotron emission model and a coherence loss of the radio emission, both for showers with zenith angle $θ\gtrsim 65^{\circ}$ and strong enough magnetic field amplitude (typical strength of $B\sim 50\, \rm μT$). Our model is compared with both ZHAireS and CoREAS Monte-Carlo simulations. Our results break the cannonical description of a radio signal made of Askaryan and transverse current emission only, and provide guidelines for the detection and reconstruction strategies of next-generation experiments, including cosmic-ray/neutrino discrimination.

astro-ph.HE↗

arXiv2024

Characterizations of $A_\infty$ Weights in Martingale Spaces

Grafakos systematically proved that $A_\infty$ weights have different characterizations for cubes in Euclidean spaces in his classical text book. Very recently, Duoandikoetxea, Mart\'ın-Reyes, Ombrosi and Kosz discussed several characterizations of the $A_{\infty}$ weights in the setting of general bases. By conditional expectations, we study $A_\infty$ weights in martingale spaces. Because conditional expectations are Radon-Nikodým derivatives with respect to sub$\hbox{-}σ\hbox{-}$fields which have no geometric structures, we need new ingredients. Under a regularity assumption on weights, we obtain equivalent characterizations of the $A_{\infty}$ weights. Moreover, using weights modulo conditional expectations, we have one-way implications of different characterizations.

math.PR↗

arXiv2024

MUBen: Benchmarking the Uncertainty of Molecular Representation Models

Large molecular representation models pre-trained on massive unlabeled data have shown great success in predicting molecular properties. However, these models may tend to overfit the fine-tuning data, resulting in over-confident predictions on test data that fall outside of the training distribution. To address this issue, uncertainty quantification (UQ) methods can be used to improve the models' calibration of predictions. Although many UQ approaches exist, not all of them lead to improved performance. While some studies have included UQ to improve molecular pre-trained models, the process of selecting suitable backbone and UQ methods for reliable molecular uncertainty estimation remains underexplored. To address this gap, we present MUBen, which evaluates different UQ methods for state-of-the-art backbone molecular representation models to investigate their capabilities. By fine-tuning various backbones using different molecular descriptors as inputs with UQ methods from different categories, we assess the influence of architectural decisions and training strategies. Our study offers insights for selecting UQ for backbone models, which can facilitate research on uncertainty-critical applications in fields such as materials science and drug discovery.

physics.chem-ph↗