arXiv subjects

Jun Han

Publications and source records attributed to Jun Han.

At least 37 records · Page 2Linked to original sources

arXiv2025

FinGAIA: A Chinese Benchmark for AI Agents in Real-World Financial Domain

The booming development of AI agents presents unprecedented opportunities for automating complex tasks across various domains. However, their multi-step, multi-tool collaboration capabilities in the financial sector remain underexplored. This paper introduces FinGAIA, an end-to-end benchmark designed to evaluate the practical abilities of AI agents in the financial domain. FinGAIA comprises 407 meticulously crafted tasks, spanning seven major financial sub-domains: securities, funds, banking, insurance, futures, trusts, and asset management. These tasks are organized into three hierarchical levels of scenario depth: basic business analysis, asset decision support, and strategic risk management. We evaluated 10 mainstream AI agents in a zero-shot setting. The best-performing agent, ChatGPT, achieved an overall accuracy of 48.9\%, which, while superior to non-professionals, still lags financial experts by over 35 percentage points. Error analysis has revealed five recurring failure patterns: Cross-modal Alignment Deficiency, Financial Terminological Bias, Operational Process Awareness Barrier, among others. These patterns point to crucial directions for future research. Our work provides the first agent benchmark closely related to the financial domain, aiming to objectively assess and promote the development of agents in this crucial field. Partial data is available at https://github.com/SUFE-AIFLM-Lab/FinGAIA.

cs.CL↗

arXiv2025

TexGS-VolVis: Expressive Scene Editing for Volume Visualization via Textured Gaussian Splatting

Advancements in volume visualization (VolVis) focus on extracting insights from 3D volumetric data by generating visually compelling renderings that reveal complex internal structures. Existing VolVis approaches have explored non-photorealistic rendering techniques to enhance the clarity, expressiveness, and informativeness of visual communication. While effective, these methods often rely on complex predefined rules and are limited to transferring a single style, restricting their flexibility. To overcome these limitations, we advocate the representation of VolVis scenes using differentiable Gaussian primitives combined with pretrained large models to enable arbitrary style transfer and real-time rendering. However, conventional 3D Gaussian primitives tightly couple geometry and appearance, leading to suboptimal stylization results. To address this, we introduce TexGS-VolVis, a textured Gaussian splatting framework for VolVis. TexGS-VolVis employs 2D Gaussian primitives, extending each Gaussian with additional texture and shading attributes, resulting in higher-quality, geometry-consistent stylization and enhanced lighting control during inference. Despite these improvements, achieving flexible and controllable scene editing remains challenging. To further enhance stylization, we develop image- and text-driven non-photorealistic scene editing tailored for TexGS-VolVis and 2D-lift-3D segmentation to enable partial editing with fine-grained control. We evaluate TexGS-VolVis both qualitatively and quantitatively across various volume rendering scenes, demonstrating its superiority over existing methods in terms of efficiency, visual quality, and editing flexibility.

cs.GR↗

arXiv2025

DTBIA: An Immersive Visual Analytics System for Brain-Inspired Research

The Digital Twin Brain (DTB) is an advanced artificial intelligence framework that integrates spiking neurons to simulate complex cognitive functions and collaborative behaviors. For domain experts, visualizing the DTB's simulation outcomes is essential to understanding complex cognitive activities. However, this task poses significant challenges due to DTB data's inherent characteristics, including its high-dimensionality, temporal dynamics, and spatial complexity. To address these challenges, we developed DTBIA, an Immersive Visual Analytics System for Brain-Inspired Research. In collaboration with domain experts, we identified key requirements for effectively visualizing spatiotemporal and topological patterns at multiple levels of detail. DTBIA incorporates a hierarchical workflow - ranging from brain regions to voxels and slice sections - along with immersive navigation and a 3D edge bundling algorithm to enhance clarity and provide deeper insights into both functional (BOLD) and structural (DTI) brain data. The utility and effectiveness of DTBIA are validated through two case studies involving with brain research experts. The results underscore the system's role in enhancing the comprehension of complex neural behaviors and interactions.

cs.HC↗

arXiv2024

SingVisio: Visual Analytics of Diffusion Model for Singing Voice Conversion

In this study, we present SingVisio, an interactive visual analysis system that aims to explain the diffusion model used in singing voice conversion. SingVisio provides a visual display of the generation process in diffusion models, showcasing the step-by-step denoising of the noisy spectrum and its transformation into a clean spectrum that captures the desired singer's timbre. The system also facilitates side-by-side comparisons of different conditions, such as source content, melody, and target timbre, highlighting the impact of these conditions on the diffusion generation process and resulting conversions. Through comparative and comprehensive evaluations, SingVisio demonstrates its effectiveness in terms of system design, functionality, explainability, and user-friendliness. It offers users of various backgrounds valuable learning experiences and insights into the diffusion model for singing voice conversion.

cs.SD↗

arXiv2024

Amphion: An Open-Source Audio, Music and Speech Generation Toolkit

Amphion is an open-source toolkit for Audio, Music, and Speech Generation, targeting to ease the way for junior researchers and engineers into these fields. It presents a unified framework that includes diverse generation tasks and models, with the added bonus of being easily extendable for new incorporation. The toolkit is designed with beginner-friendly workflows and pre-trained models, allowing both beginners and seasoned researchers to kick-start their projects with relative ease. The initial release of Amphion v0.1 supports a range of tasks including Text to Speech (TTS), Text to Audio (TTA), and Singing Voice Conversion (SVC), supplemented by essential components like data preprocessing, state-of-the-art vocoders, and evaluation metrics. This paper presents a high-level overview of Amphion. Amphion is open-sourced at https://github.com/open-mmlab/Amphion.

cs.SD↗

arXiv2024

Contextual Object Detection with Multimodal Large Language Models

Recent Multimodal Large Language Models (MLLMs) are remarkable in vision-language tasks, such as image captioning and question answering, but lack the essential perception ability, i.e., object detection. In this work, we address this limitation by introducing a novel research problem of contextual object detection -- understanding visible objects within different human-AI interactive contexts. Three representative scenarios are investigated, including the language cloze test, visual captioning, and question answering. Moreover, we present ContextDET, a unified multimodal model that is capable of end-to-end differentiable modeling of visual-language contexts, so as to locate, identify, and associate visual objects with language inputs for human-AI interaction. Our ContextDET involves three key submodels: (i) a visual encoder for extracting visual representations, (ii) a pre-trained LLM for multimodal context decoding, and (iii) a visual decoder for predicting bounding boxes given contextual object words. The new generate-then-detect framework enables us to detect object words within human vocabulary. Extensive experiments show the advantages of ContextDET on our proposed CODE benchmark, open-vocabulary detection, and referring image segmentation. Github: https://github.com/yuhangzang/ContextDET.

cs.CV↗

arXiv2024

Guided Discrete Diffusion for Electronic Health Record Generation

Electronic health records (EHRs) are a pivotal data source that enables numerous applications in computational medicine, e.g., disease progression prediction, clinical trial design, and health economics and outcomes research. Despite wide usability, their sensitive nature raises privacy and confidentially concerns, which limit potential use cases. To tackle these challenges, we explore the use of generative models to synthesize artificial, yet realistic EHRs. While diffusion-based methods have recently demonstrated state-of-the-art performance in generating other data modalities and overcome the training instability and mode collapse issues that plague previous GAN-based approaches, their applications in EHR generation remain underexplored. The discrete nature of tabular medical code data in EHRs poses challenges for high-quality data generation, especially for continuous diffusion models. To this end, we introduce a novel tabular EHR generation method, EHR-D3PM, which enables both unconditional and conditional generation using the discrete diffusion model. Our experiments demonstrate that EHR-D3PM significantly outperforms existing generative baselines on comprehensive fidelity and utility metrics while maintaining less attribute and membership vulnerability risks. Furthermore, we show EHR-D3PM is effective as a data augmentation method and enhances performance on downstream tasks when combined with real data.

cs.LG↗

arXiv2022

Photometric redshift estimation of galaxies in the DESI Legacy Imaging Surveys

The accurate estimation of photometric redshifts plays a crucial role in accomplishing science objectives of the large survey projects. The template-fitting and machine learning are the two main types of methods applied currently. Based on the training set obtained by cross-correlating the DESI Legacy Imaging Surveys DR9 galaxy catalogue and SDSS DR16 galaxy catalogue, the two kinds of methods are used and optimized, such as EAZY for template-fitting approach and CATBOOST for machine learning. Then the created models are tested by the cross-matched samples of the DESI Legacy Imaging SurveysDR9 galaxy catalogue with LAMOST DR7, GAMA DR3 and WiggleZ galaxy catalogues. Moreover three machine learning methods (CATBOOST, Multi-Layer Perceptron and Random Forest) are compared, CATBOOST shows its superiority for our case. By feature selection and optimization of model parameters, CATBOOST can obtain higher accuracy with optical and infrared photometric information, the best performance ($MSE=0.0032$, $σ_{NMAD}=0.0156$ and $O=0.88$ per cent) with $g \le 24.0$, $r \le 23.4$ and $z \le 22.5$ is achieved. But EAZY can provide more accurate photometric redshift estimation for high redshift galaxies, especially beyond the redhisft range of training sample. Finally, we finish the redshift estimation of all DESI DR9 galaxies with CATBOOST and EAZY, which will contribute to the further study of galaxies and their properties.

astro-ph.IM↗

arXiv2022

Projected Gradient Descent Algorithms for Solving Nonlinear Inverse Problems with Generative Priors

In this paper, we propose projected gradient descent (PGD) algorithms for signal estimation from noisy nonlinear measurements. We assume that the unknown $p$-dimensional signal lies near the range of an $L$-Lipschitz continuous generative model with bounded $k$-dimensional inputs. In particular, we consider two cases when the nonlinear link function is either unknown or known. For unknown nonlinearity, similarly to \cite{liu2020generalized}, we make the assumption of sub-Gaussian observations and propose a linear least-squares estimator. We show that when there is no representation error and the sensing vectors are Gaussian, roughly $O(k \log L)$ samples suffice to ensure that a PGD algorithm converges linearly to a point achieving the optimal statistical rate using arbitrary initialization. For known nonlinearity, we assume monotonicity as in \cite{yang2016sparse}, and make much weaker assumptions on the sensing vectors and allow for representation error. We propose a nonlinear least-squares estimator that is guaranteed to enjoy an optimal statistical rate. A corresponding PGD algorithm is provided and is shown to also converge linearly to the estimator using arbitrary initialization. In addition, we present experimental results on image datasets to demonstrate the performance of our PGD algorithms.

stat.ML↗

arXiv2022

Generative Principal Component Analysis

In this paper, we study the problem of principal component analysis with generative modeling assumptions, adopting a general model for the observed matrix that encompasses notable special cases, including spiked matrix recovery and phase retrieval. The key assumption is that the underlying signal lies near the range of an $L$-Lipschitz continuous generative model with bounded $k$-dimensional inputs. We propose a quadratic estimator, and show that it enjoys a statistical rate of order $\sqrt{\frac{k\log L}{m}}$, where $m$ is the number of samples. We also provide a near-matching algorithm-independent lower bound. Moreover, we provide a variant of the classic power method, which projects the calculated data onto the range of the generative model during each iteration. We show that under suitable conditions, this method converges exponentially fast to a point achieving the above-mentioned statistical rate. We perform experiments on various image datasets for spiked matrix and phase retrieval models, and illustrate performance gains of our method to the classic power method and the truncated power method devised for sparse principal component analysis.

stat.ML↗

arXiv2022

TickTock: Detecting Microphone Status in Laptops Leveraging Electromagnetic Leakage of Clock Signals

We are witnessing a heightened surge in remote privacy attacks on laptop computers. These attacks often exploit malware to remotely gain access to webcams and microphones in order to spy on the victim users. While webcam attacks are somewhat defended with widely available commercial webcam privacy covers, unfortunately, there are no adequate solutions to thwart the attacks on mics despite recent industry efforts. As a first step towards defending against such attacks on laptop mics, we propose TickTock, a novel mic on/off status detection system. To achieve this, TickTock externally probes the electromagnetic (EM) emanations that stem from the connectors and cables of the laptop circuitry carrying mic clock signals. This is possible because the mic clock signals are only input during the mic recording state, causing resulting emanations. We design and implement a proof-of-concept system to demonstrate TickTock's feasibility. Furthermore, we comprehensively evaluate TickTock on a total of 30 popular laptops executing a variety of applications to successfully detect mic status in 27 laptops. Of these, TickTock consistently identifies mic recording with high true positive and negative rates.

cs.CR↗

arXiv2022

DL4SciVis: A State-of-the-Art Survey on Deep Learning for Scientific Visualization

Since 2016, we have witnessed the tremendous growth of artificial intelligence+visualization (AI+VIS) research. However, existing survey papers on AI+VIS focus on visual analytics and information visualization, not scientific visualization (SciVis). In this paper, we survey related deep learning (DL) works in SciVis, specifically in the direction of DL4SciVis: designing DL solutions for solving SciVis problems. To stay focused, we primarily consider works that handle scalar and vector field data but exclude mesh data. We classify and discuss these works along six dimensions: domain setting, research task, learning type, network architecture, loss function, and evaluation metric. The paper concludes with a discussion of the remaining gaps to fill along the discussed dimensions and the grand challenges we need to tackle as a community. This state-of-the-art survey guides SciVis researchers in gaining an overview of this emerging topic and points out future directions to grow this research.

cs.GR↗

arXiv2022

FastZIP: Faster and More Secure Zero-Interaction Pairing

With the advent of the Internet of Things (IoT), establishing a secure channel between smart devices becomes crucial. Recent research proposes zero-interaction pairing (ZIP), which enables pairing without user assistance by utilizing devices' physical context (e.g., ambient audio) to obtain a shared secret key. The state-of-the-art ZIP schemes suffer from three limitations: (1) prolonged pairing time (i.e., minutes or hours), (2) vulnerability to brute-force offline attacks on a shared key, and (3) susceptibility to attacks caused by predictable context (e.g., replay attack) because they rely on limited entropy of physical context to protect a shared key. We address these limitations, proposing FastZIP, a novel ZIP scheme that significantly reduces pairing time while preventing offline and predictable context attacks. In particular, we adapt a recently introduced Fuzzy Password-Authenticated Key Exchange (fPAKE) protocol and utilize sensor fusion, maximizing their advantages. We instantiate FastZIP for intra-car device pairing to demonstrate its feasibility and show how the design of FastZIP can be adapted to other ZIP use cases. We implement FastZIP and evaluate it by driving four cars for a total of 800 km. We achieve up to three times shorter pairing time compared to the state-of-the-art ZIP schemes while assuring robust security with adversarial error rates below 0.5%.

cs.CR↗

arXiv2022

Non-invasive chemically selective energy delivery and focusing inside a scattering medium guided by Raman scattering

Raman scattering is a chemically selective probing mechanism with diverse applications in industry and clinical settings. Yet, most samples are optically opaque limiting the applicability of Raman probing at depth. Here, we demonstrate chemically selective energy deposition behind a scattering medium by combining prior information on the chemical's spectrum with the measurement of a spectrally resolved Raman speckle as a feedback mechanism for wavefront shaping. We demonstrate unprecedented six-fold signal enhancement in an epi-geometry, realizing targeted energy deposition and focusing on selected Raman active particles.

physics.optics↗

arXiv2021

Photometric Redshift Estimation of BASS DR3 Quasars by Machine Learning

Correlating BASS DR3 catalogue with ALLWISE database, the data from optical and infrared information are obtained. The quasars from SDSS are taken as training and test samples while those from LAMOST are considered as external test sample. We propose two schemes to construct the redshift estimation models with XGBoost, CatBoost and Random forest. One scheme (namely one-step model) is to predict photometric redshifts directly based on the optimal models created by these three algorithms; the other scheme (namely two-step model) is to firstly classify the data into low- and high- redshift datasets, and then predict photometric redshifts of these two datasets separately. For one-step model, the performance of these three algorithms on photometric redshift estimation is compared with different training samples, and CatBoost is superior to XGBoost and Random forest. For two-step model, the performance of these three algorithms on the classification of low- and high-redshift subsamples are compared, and CatBoost still shows the best performance. Therefore CatBoost is regard as the core algorithm of classification and regression in two-step model. By contrast with one-step model, two-step model is optimal when predicting photometric redshift of quasars, especially for high redshift quasars. Finally the two models are applied to predict photometric redshifts of all quasar candidates of BASS DR3. The number of high redshift quasar candidates is 3938 (redshift $\ge 3.5$) and 121 (redshift $\ge 4.5$) by two-step model. The predicted result will be helpful for quasar research and follow up observation of high redshift quasars.

astro-ph.IM↗

arXiv2021

Touchtone leakage attacks via smartphone sensors: mitigation without hardware modification

Smartphone motion sensors provide a concealed mechanism for eavesdropping on acoustic information, like touchtones, emitted by a device. Eavesdropping on touchtones exposes credit card information, banking pins, and social security card numbers to malicious 3rd party apps requiring only motion sensor data. This paper's primary contribution is an analysis rooted in physics and signal processing theory of several eavesdropping mitigations, which could be implemented in a smartphone update. We verify our analysis imperially to show how previously suggested mitigations, i.e. a low-pass filter, can undesirably reduce the motion sensor data to all applications by 83% but only reduce an advanced adversary's accuracy by less than one percent. Other designs, i.e. anti-aliasing filters, can fully preserve the motion sensor data to support benign application functionality while reducing attack accuracy by 50.1%. We intend for this analysis to motivate the need for deployable mitigations against acoustic leakage on smartphone motion sensors, including but not limited to touchtones, while also providing a basis for future mitigations to improve upon.

cs.CR↗

arXiv2021

Robust 1-bit Compressive Sensing with Partial Gaussian Circulant Matrices and Generative Priors

In 1-bit compressive sensing, each measurement is quantized to a single bit, namely the sign of a linear function of an unknown vector, and the goal is to accurately recover the vector. While it is most popular to assume a standard Gaussian sensing matrix for 1-bit compressive sensing, using structured sensing matrices such as partial Gaussian circulant matrices is of significant practical importance due to their faster matrix operations. In this paper, we provide recovery guarantees for a correlation-based optimization algorithm for robust 1-bit compressive sensing with randomly signed partial Gaussian circulant matrices and generative models. Under suitable assumptions, we match guarantees that were previously only known to hold for i.i.d.~Gaussian matrices that require significantly more computation. We make use of a practical iterative algorithm, and perform numerical experiments on image datasets to corroborate our theoretical results.

cs.LG↗

arXiv2021

TS4Net: Two-Stage Sample Selective Strategy for Rotating Object Detection

Rotating object detection has wide applications in aerial photographs, remote sensing images, UAVs, etc. At present, most of the rotating object detection datasets focus on the field of remote sensing, and these images are usually shot in high-altitude scenes. However, image datasets captured at low-altitude areas also should be concerned, such as drone-based datasets. So we present a low-altitude dronebased dataset, named UAV-ROD, aiming to promote the research and development in rotating object detection and UAV applications. The UAV-ROD consists of 1577 images and 30,090 instances of car category annotated by oriented bounding boxes. In particular, The UAV-ROD can be utilized for the rotating object detection, vehicle orientation recognition and object counting tasks. Compared with horizontal object detection, the regression stage of the rotation detection is a tricky problem. In this paper, we propose a rotating object detector TS4Net, which contains anchor refinement module (ARM) and two-stage sample selective strategy (TS4). The ARM can convert preseted horizontal anchors into high-quality rotated anchors through twostage anchor refinement. The TS4 module utilizes different constrained sample selective strategies to allocate positive and negative samples, which is adaptive to the regression task in different stages. Benefiting from the ARM and TS4, the TS4Net can achieve superior performance for rotating object detection solely with one preseted horizontal anchor. Extensive experimental results on UAV-ROD dataset and three remote sensing datasets DOTA, HRSC2016 and UCAS-AOD demonstrate that our method achieves competitive performance against most state-of-the-art methods.

cs.CV↗