arXiv subjects

Shan Zhong

Publications and source records attributed to Shan Zhong.

At least 19 recordsLinked to original sources

arXiv2026

AgenticDataBench: A Comprehensive Benchmark for Data Agents

Data science aims to derive actionable insights from heterogeneous raw data, unlocking the value of the massive amounts of data generated in modern society. Automating this process is essential to reducing labor-intensive efforts for data scientists and enabling scalable data-driven applications. Recently, large language model (LLM)-based data agents have emerged as a promising solution to automate data science workflows. However, the field lacks comprehensive benchmarks to rigorously evaluate these agents across diverse scenarios with fine-grained granularity. To address this gap, we propose AgenticDataBench, a comprehensive benchmark featuring realistic tasks spanning diverse domains with fine-grained ground-truth labels. This enables evaluations to capture the diversity and complexity of data science workflows and the detailed performance of agents. First, to cover diverse domains, we collect real datasets and tasks from 15 vertical domains, including 5 real-world B2B use cases from a leading fintech company. Second, to remove redundancy in real-world tasks and generate high-quality tasks for domains lacking real data, we introduce data science skills, recurring data-centric operational patterns, and quantify benchmark coverage by the number of skills included. Representative skills are extracted from large-scale task solutions on Stack Overflow using skill-aligned hierarchical clustering. Third, for real-world business tasks, we select task-solution pairs that maximize diversity in skill composition, ensuring broad coverage of practical scenarios. Fourth, to generate realistic tasks for devise domains without real tasks, we propose a systematic LLM-based task generation approach to create workflows and tasks based on these skills. Finally, we evaluate state-of-the-art data agents using our annotated benchmark and open-sourced testbed, providing detailed skill-level insights.

cs.DB↗

arXiv2026

VesNet: Neural network accelerated solver for simulating Stokesian vesicle suspensions

Numerical simulation of deformable particle suspensions in Stokes flow is computationally expensive due to nonlinear fluid-structure interactions, evolving interfaces, and multiscale hydrodynamics. We present VesNet, a hybrid framework that accelerates two-dimensional vesicle suspension simulations by approximating vesicle self interactions, including background flow coupling and short-range lubrication forces, while retaining conventional modules for boundary reparameterization and far-field hydrodynamics. A GPU-accelerated implementation achieves over 100x speedup compared to a multithreaded MATLAB CPU boundary integral solver and about 5x relative to its GPU counterpart. VesNet accurately captures key dynamics, including single-vesicle phase behavior, pair interactions, and large-scale suspensions in Taylor-Green and Poiseuille flows, enabling efficient simulations of thousands of vesicles on modest computational resources.

physics.flu-dyn↗

arXiv2026

IV-Net: A neural network for elliptic PDEs with random and highly varying coefficients

We introduce a novel neural operator architecture designed to approximate solutions of linear elliptic partial differential equations with high-contrast, spatially varying coefficients. The network, termed the Iterated V-shaped Net (IV-Net), realizes a mapping from the input coefficients and righthand side to the corresponding solution field. The architecture of IV-Net is informed by, and closely resembles, a V-cycle multigrid solver. The IV-Net model is parameterized via convolutional layers defined in the physical domain. For coercive problems with highly heterogeneous coefficients, the proposed network exhibits superior performance relative to a proper orthogonal decomposition (POD) approach and several existing neural operator architectures. For low-frequency oscillatory Helmholtz problems with smooth coefficients, its performance is similar to that of a Fourier neural operator. We analyze the approximation error and convergence behavior of IV-Net, its data efficiency, and its dependence on the underlying discretization mesh. Furthermore, we demonstrate the practical effectiveness of the architecture through a series of numerical experiments, including applications to uncertainty quantification, inverse problems, and prediction of quantities of interest.

math.NA↗

arXiv2026

Distributional Reinforcement Learning with Diffusion Bridge Critics

Recent advances in diffusion-based reinforcement learning (RL) methods have demonstrated promising results in a wide range of continuous control tasks. However, existing works in this field focus on the application of diffusion policies while leaving the diffusion critics unexplored. In fact, since policy optimization fundamentally relies on the critic, accurate value estimation is far more important than policy expressiveness. Furthermore, given the stochasticity of most reinforcement learning tasks, it has been confirmed that the critic is more appropriately depicted with a distributional model. Motivated by these points, we propose a novel distributional RL method with Diffusion Bridge Critics (DBC). DBC directly models the inverse cumulative distribution function (CDF) of the Q value. This allows us to accurately capture the value distribution and prevents it from collapsing into a trivial Gaussian distribution owing to the strong distribution-matching capability of the diffusion bridge. Moreover, we further derive an analytic integral formula to address discretization errors in DBC, which is essential in value estimation. To our knowledge, DBC is the first work to employ the diffusion bridge model as the critic. Notably, DBC is also a plug-and-play component and can be integrated into most existing RL frameworks. Experimental results on MuJoCo robot control benchmarks demonstrate the superiority of DBC compared with previous distributional critic models.

cs.LG↗

arXiv2026

GenPO: Generative Diffusion Models Meet On-Policy Reinforcement Learning

Recent advances in reinforcement learning (RL) have demonstrated the powerful exploration capabilities and multimodality of generative diffusion-based policies. While substantial progress has been made in offline RL and off-policy RL settings, integrating diffusion policies into on-policy frameworks like PPO remains underexplored. This gap is particularly significant given the widespread use of large-scale parallel GPU-accelerated simulators, such as IsaacLab, which are optimized for on-policy RL algorithms and enable rapid training of complex robotic tasks. A key challenge lies in computing state-action log-likelihoods under diffusion policies, which is straightforward for Gaussian policies but intractable for flow-based models due to irreversible forward-reverse processes and discretization errors (e.g., Euler-Maruyama approximations). To bridge this gap, we propose GenPO, a generative policy optimization framework that leverages exact diffusion inversion to construct invertible action mappings. GenPO introduces a novel doubled dummy action mechanism that enables invertibility via alternating updates, resolving log-likelihood computation barriers. Furthermore, we also use the action log-likelihood for unbiased entropy and KL divergence estimation, enabling KL-adaptive learning rates and entropy regularization in on-policy updates. Extensive experiments on eight IsaacLab benchmarks, including legged locomotion (Ant, Humanoid, Anymal-D, Unitree H1, Go2), dexterous manipulation (Shadow Hand), aerial control (Quadcopter), and robotic arm tasks (Franka), demonstrate GenPO's superiority over existing RL baselines. Notably, GenPO is the first method to successfully integrate diffusion policies into on-policy RL, unlocking their potential for large-scale parallelized training and real-world robotic deployment.

cs.LG↗

arXiv2025

Deep Reinforcement Learning for Automated Stock Trading: An Ensemble Strategy

Stock trading strategies play a critical role in investment. However, it is challenging to design a profitable strategy in a complex and dynamic stock market. In this paper, we propose an ensemble strategy that employs deep reinforcement schemes to learn a stock trading strategy by maximizing investment return. We train a deep reinforcement learning agent and obtain an ensemble trading strategy using three actor-critic based algorithms: Proximal Policy Optimization (PPO), Advantage Actor Critic (A2C), and Deep Deterministic Policy Gradient (DDPG). The ensemble strategy inherits and integrates the best features of the three algorithms, thereby robustly adjusting to different market situations. In order to avoid the large memory consumption in training networks with continuous action space, we employ a load-on-demand technique for processing very large data. We test our algorithms on the 30 Dow Jones stocks that have adequate liquidity. The performance of the trading agent with different reinforcement learning algorithms is evaluated and compared with both the Dow Jones Industrial Average index and the traditional min-variance portfolio allocation strategy. The proposed deep ensemble strategy is shown to outperform the three individual algorithms and two baselines in terms of the risk-adjusted return measured by the Sharpe ratio. This work is fully open-sourced at \href{https://github.com/AI4Finance-Foundation/Deep-Reinforcement-Learning-for-Automated-Stock-Trading-Ensemble-Strategy-ICAIF-2020}{GitHub}.

q-fin.TR↗

arXiv2025

FlowCritic: Bridging Value Estimation with Flow Matching in Reinforcement Learning

Reliable value estimation serves as the cornerstone of reinforcement learning (RL) by evaluating long-term returns and guiding policy improvement, significantly influencing the convergence speed and final performance. Existing works improve the reliability of value function estimation via multi-critic ensembles and distributional RL, yet the former merely combines multi point estimation without capturing distributional information, whereas the latter relies on discretization or quantile regression, limiting the expressiveness of complex value distributions. Inspired by flow matching's success in generative modeling, we propose a generative paradigm for value estimation, named FlowCritic. Departing from conventional regression for deterministic value prediction, FlowCritic leverages flow matching to model value distributions and generate samples for value estimation.

cs.LG↗

arXiv2025

Packaging of Low-Environmental-Sensitivity WGM Resonators for Practical Applications

We present a prism-coupled packaging strategy for whispering-gallery mode resonators (WGMRs). Utilizing an all-solid-state optical adhesive process with active temperature control and hermetic sealing, the package exhibits exceptional long-term stability and environmental robustness. A standalone WGMR module was characterized, demonstrating a temperature sensitivity below 10^-7 /°C and a low-frequency Z-axis acceleration sensitivity below 10^-10 /g. The module was applied as a stable optical frequency reference, achieving a short-term frequency stability of 2x10^-13 at 2 ms, and as a nonlinear photonic platform, generating Kerr soliton microcombs with a pump power of 100 mW. This compact, robust packaging solution enhances the applicability of WGMRs in real-world applications such as narrow-linewidth lasers and portable microcombs, facilitating the transition from laboratory research to practical deployment.

physics.optics↗

arXiv2025

Chip-Scale Rydberg Atomic Electrometer

An ideal electrometer should measure electric fields accurately while causing minimal disturbance to the field itself. Rydberg atomic electrometers are promising candidates for ideal electrometry due to their SI traceability and non-invasive nature. However, in practice, the atomic vapor cell shell can distort the electric field, limiting the device's performance. In this work, we overcome this challenge by fabricating a chip-scale vapor cell using a novel combination of femtosecond laser writing and optical contact. This method enables the development of a non-invasive atomic electrometer with a radar cross-section (RCS) 20 dB lower than that of commercial atomic cell-based electrometers. Furthermore, we observe a new sub-Doppler spectral narrowing phenomenon in these chip-scale cells. The effect originates from an incoherent, collision-driven mechanism--hereafter referred to as incoherent Dicke narrowing (ICDN). This advancement supports future revisions to the international system of units and broadens applications in metrology and quantum measurement.

physics.atom-ph↗

arXiv2025

Functional Ultrasound Imaging Combined with Machine Learning for Whole-Brain Analysis of Drug-Induced Hemodynamic Changes

Functional ultrasound imaging (fUSI) is a cutting-edge technology that measures changes in cerebral blood volume (CBV) by detecting backscattered echoes from red blood cells moving within its field of view (FOV). It offers high spatiotemporal resolution and sensitivity, allowing for detailed visualization of cerebral blood flow dynamics. While fUSI has been utilized in preclinical drug development studies to explore the mechanisms of action of various drugs targeting the central nervous system, many of these studies rely on predetermined regions of interest (ROIs). This focus may overlook relevant brain activity outside these specific areas, which could influence the results. To address this limitation, we compared three machine learning approaches-convolutional neural network (CNN), support vector machine (SVM), and vision transformer (ViT)-combined with fUSI to analyze the pharmacodynamics of Dizocilpine (MK-801), a potent non-competitive NMDA receptor antagonist commonly used in preclinical models for memory and learning impairments. While all three machine learning techniques could distinguish between drug and control conditions, CNN proved particularly effective due to their ability to capture hierarchical spatial features while maintaining anatomical specificity. Class activation mapping revealed brain regions, including the prefrontal cortex and hippocampus, that are significantly affected by drug administration, consistent with the literature reporting a high density of NMDA receptors in these areas. Overall, the combination of fUSI and CNN creates a novel analytical framework for examining pharmacological mechanisms, allowing for data-driven identification and regional mapping of drug effects while preserving anatomical context and physiological relevance.

q-bio.NC↗

arXiv2025

Light Storage and Retrieval in an Atomic Tripod System

Highly-efficient quantum memories are essential for advancing quantum information processing technologies, including scalable quantum computing and quantum networks. We experimentally demonstrate a light storage and retrieval protocol in a tripod system using an ensemble of laser-cooled $^{87}$Rb atoms. The tripod system, which consists of three ground states and an excited state, offers rich dynamics: its use to coherently store and retrieve a weak probe pulse in the $^{87}$Rb $F=1$ ground state manifold leads to the interference of two spin-wave excitations during storage time that translate to an interference in the peak intensity of the retrieved probe pulse. Our work shows that these interferences, which manifest when varying the pulse sequence or energy level structure, can be controlled experimentally by varying the storage time, optical phase, and magnetic field strength. Theoretical simulations exhibit excellent agreement with the experimental results. This work demonstrates the rich dynamics and versatile capabilities of atomic tripod systems for light storage and retrieval, with key advantages over conventional $Λ$-systems, highlighting the potential of atomic tripod systems for applications in quantum information processing, quantum synchronization, and atomic memory protocols.

quant-ph↗

arXiv2024

Clustering Algorithms and RAG Enhancing Semi-Supervised Text Classification with Large LLMs

This paper proposes a Clustering, Labeling, then Augmenting framework that significantly enhances performance in Semi-Supervised Text Classification (SSTC) tasks, effectively addressing the challenge of vast datasets with limited labeled examples. Unlike traditional SSTC approaches that rely on a predefined small set of labeled data to generate pseudo-labels for the unlabeled data, this framework innovatively employs clustering to select representative "landmarks" for labeling. These landmarks subsequently act as intermediaries in an ensemble of augmentation techniques, including Retrieval-Augmented Generation (RAG), Large Language Model (LLMs)-based rewriting, and synonym substitution, to generate synthetic labeled data without making pseudo-labels for the unlabeled data. Empirical results show that even in complex text document classification scenarios involving over 100 categories, our method achieves state-of-the-art accuracies of 95.41% on the Reuters dataset and 82.43% on the Web of Science dataset. Our approach significantly reduces the reliance on human labeling efforts and the associated expenses, while simultaneously ensuring high data quality and minimizing privacy risks. The finetuning results further show the efficiency of fine-tuning LLMs for text classification tasks, highlighting a robust solution for leveraging limited labeled data.

cs.CL↗

arXiv2024

Adaptive Environment-Aware Robotic Arm Reaching Based on a Bio-Inspired Neurodynamical Computational Framework

Bio-inspired robotic systems are capable of adaptive learning, scalable control, and efficient information processing. Enabling real-time decision-making for such systems is critical to respond to dynamic changes in the environment. We focus on dynamic target tracking in open areas using a robotic six-degree-of-freedom manipulator with a bird-eye view camera for visual feedback, and by deploying the Neurodynamical Computational Framework (NeuCF). NeuCF is a recently developed bio-inspired model for target tracking based on Dynamic Neural Fields (DNFs) and Stochastic Optimal Control (SOC) theory. It has been trained for reaching actions on a planar surface toward localized visual beacons, and it can re-target or generate stop signals on the fly based on changes in the environment (e.g., a new target has emerged, or an existing one has been removed). We evaluated our system over various target-reaching scenarios. In all experiments, NeuCF had high end-effector positional accuracy, generated smooth trajectories, and provided reduced path lengths compared with a baseline cubic polynomial trajectory generator. In all, the developed system offers a robust and dynamic-aware robotic manipulation approach that affords real-time decision-making.

cs.RO↗

arXiv2024

Differentially Private Low-Rank Adaptation of Large Language Model Using Federated Learning

The surge in interest and application of large language models (LLMs) has sparked a drive to fine-tune these models to suit specific applications, such as finance and medical science. However, concerns regarding data privacy have emerged, especially when multiple stakeholders aim to collaboratively enhance LLMs using sensitive data. In this scenario, federated learning becomes a natural choice, allowing decentralized fine-tuning without exposing raw data to central servers. Motivated by this, we investigate how data privacy can be ensured in LLM fine-tuning through practical federated learning approaches, enabling secure contributions from multiple parties to enhance LLMs. Yet, challenges arise: 1) despite avoiding raw data exposure, there is a risk of inferring sensitive information from model outputs, and 2) federated learning for LLMs incurs notable communication overhead. To address these challenges, this article introduces DP-LoRA, a novel federated learning algorithm tailored for LLMs. DP-LoRA preserves data privacy by employing a Gaussian mechanism that adds noise in weight updates, maintaining individual data privacy while facilitating collaborative model training. Moreover, DP-LoRA optimizes communication efficiency via low-rank adaptation, minimizing the transmission of updated weights during distributed training. The experimental results across medical, financial, and general datasets using various LLMs demonstrate that DP-LoRA effectively ensures strict privacy constraints while minimizing communication overhead.

cs.LG↗

arXiv2024

WGMR Self-Injection Locking Method Based on Enhanced Optical Feedback with Auxiliary Prism

The optical feedback intensity is an important parameter for realizing narrow linewidth lasers in Whispering-gallery-mode resonator (WGMR) self-injection locking technology. We proposed an approach that enhances the intensity of intracavity feedback in crystalline WGMR by using only a single coated auxiliary prism. Compared to the Rayleigh scattering, the feedback intensity of the enhanced scheme increased by more than a hundred times. Furthermore, we demonstrated that, with the enhanced approach, the instantaneous linewidth of the laser was suppressed to 7 Hz level, the locking range was expanded up to 8 GHz, and the relative intensity noise (RIN) was reduced to -152 dBc/Hz@10MHz. The feedback enhanced design is compact, easy-to-operated and can be integrated with the WGMR. It provides a miniaturized solution for controlling optical feedback intensity in WGMR self-injection locking technology.

physics.optics↗

arXiv2023

Variational Bayesian Approximations Kalman Filter Based on Threshold Judgment

The estimation of non-Gaussian measurement noise models is a significant challenge across various fields. In practical applications, it often faces challenges due to the large number of parameters and high computational complexity. This paper proposes a threshold-based Kalman filtering approach for online estimation of noise parameters in non-Gaussian measurement noise models. This method uses a certain amount of sample data to infer the variance threshold of observation parameters and employs variational Bayesian estimation to obtain corresponding noise variance estimates, enabling subsequent iterations of the Kalman filtering algorithm. Finally, we evaluate the performance of this algorithm through simulation experiments, demonstrating its accurate and effective estimation of state and noise parameters.

eess.SY↗

arXiv2023

Interactive Model Fusion-Based GM-PHD Filter

In multi-target tracking (MTT), non-Gaussian measurement noise from sensors can diminish the performance of the Gaussian-assumed Gaussian mixture probability hypothesis density (GM-PHD) filter. In this paper, an approach that transforms the MTT problem under non-Gaussian conditions into an MTT problem under Gaussian conditions is developed. Specifically, measurement noise with a non-Gaussian distribution is modeled as a weighted sum of different Gaussian distributions. Subsequently, the GM-PHD filter is applied to compute the multi-target states under these distinct Gaussian distributions. Finally, an interactive multi-model framework is employed to fuse the diverse multi-target state information into a unified synthesis. The effectiveness of the proposed approach is validated through the simulation results.

eess.SY↗

arXiv2023

Quantized criterion-based kernel recursive least squares adaptive filtering for time series prediction

The robustness of the kernel recursive least square (KRLS) algorithm has recently been improved by combining them with more robust information-theoretic learning criteria, such as minimum error entropy (MEE) and generalized MEE (GMEE), which also improves the computational complexity of the KRLS-type algorithms to a certain extent. To reduce the computational load of the KRLS-type algorithms, the quantized GMEE (QGMEE) criterion, in this paper, is combined with the KRLS algorithm, and as a result two kinds of KRLS-type algorithms, called quantized kernel recursive MEE (QKRMEE) and quantized kernel recursive GMEE (QKRGMEE), are designed. As well, the mean error behavior, mean square error behavior, and computational complexity of the proposed algorithms are investigated. In addition, simulation and real experimental data are utilized to verify the feasibility of the proposed algorithms.

cs.IT↗