arXiv subjects

Xin Zhang

Publications and source records attributed to Xin Zhang.

At least 523 records · Page 29Linked to original sources

arXiv2025

FMDLlama: Financial Misinformation Detection based on Large Language Models

The emergence of social media has made the spread of misinformation easier. In the financial domain, the accuracy of information is crucial for various aspects of financial market, which has made financial misinformation detection (FMD) an urgent problem that needs to be addressed. Large language models (LLMs) have demonstrated outstanding performance in various fields. However, current studies mostly rely on traditional methods and have not explored the application of LLMs in the field of FMD. The main reason is the lack of FMD instruction tuning datasets and evaluation benchmarks. In this paper, we propose FMDLlama, the first open-sourced instruction-following LLMs for FMD task based on fine-tuning Llama3.1 with instruction data, the first multi-task FMD instruction dataset (FMDID) to support LLM instruction tuning, and a comprehensive FMD evaluation benchmark (FMD-B) with classification and explanation generation tasks to test the FMD ability of LLMs. We compare our models with a variety of LLMs on FMD-B, where our model outperforms other open-sourced LLMs as well as OpenAI's products. This project is available at https://github.com/lzw108/FMD.

cs.CL↗

arXiv2025

Decentralized MIMO Systems with Imperfect CSI using LMMSE Receivers

Centralized baseband processing (CBP) is required to achieve the full potential of massive multiple-input multiple-output (MIMO) systems. However, due to the large number of antennas, CBP suffers from two major issues: 1) Tremendous data interconnection between radio frequency (RF) circuitry and processing fabrics; and 2) high-dimensional computation. To this end, decentralized baseband processing (DBP) has been proposed, where the antennas at the BS are partitioned into clusters connected to separate RF circuits and equipped with separate computing units. Unfortunately, due to the decentralized structure, the optimal fusion scheme and performance analysis for DBP with general spatial correlation between clusters and imperfect channel state information (CSI) are not available in the literature. In this paper, we consider a decentralized MIMO system where all clusters adopt linear minimum mean-square error (LMMSE) receivers with imperfect CSI. Specifically, we first establish the optimal linear fusion scheme which has high computational and data input/output (I/O) costs. To reduce the costs, we further propose two sub-optimal fusion schemes with reduced complexity. For all three schemes, we derive the closed-form expressions for the signal-to-interference-and-noise ratio (SINR) by leveraging random matrix theory (RMT) and demonstrate the conditions under which the sub optimal schemes are optimal. Furthermore, we determine the optimal regularization parameter for decentralized LMMSE receivers, identify the best antenna partitioning strategy, and prove that the SINR will decrease as the number of clusters increases. Numerical simulations validate the accuracy of the theoretical results.

cs.IT↗

arXiv2025

Geometric properties of the additional third-order transitions in the two-dimensional Potts model

Within the canonical ensemble framework, this paper investigates the presence of higher-order transition signals in the $q$-state Potts model (for $q \geq 3$), using two geometric order parameters: isolated spins number and the average perimeter of clusters. Our results confirm that higher-order transitions exist in the Potts model, where the number of isolated spins reliably indicates third-order independent transitions. This signal persists regardless of the system's phase transition order, even at higher values of $q$. In contrast, the average perimeter of clusters, used as an order parameter for detecting third-order dependent transitions, shows that for $q = 6$ and $q = 8$, the signal for third-order dependent transitions disappears, indicating its absence in systems undergoing first-order transitions. These findings are consistent with results from microcanonical inflection-point analysis, further validating the robustness of this approach.

cond-mat.stat-mech↗

arXiv2025

The Wasserstein Space of Stochastic Processes in Continuous Time

Researchers from different areas have independently defined extensions of the usual weak convergence of laws of stochastic processes with the goal of adequately accounting for the flow of information. Natural approaches are convergence of the Aldous--Knight prediction process, Hellwig's information topology, convergence in adapted distribution in the sense of Hoover--Keisler and the weak topology induced by optimal stopping problems. The first main contribution of this article is that on continuous processes with natural filtrations there exists a canonical adapted weak topology which can be defined by all of these approaches; moreover, the adapted weak topology is metrized by a suitable adapted Wasserstein distance $\mathcal{AW}$. While the set of processes with natural filtrations is not complete, we establish that its completion consists precisely of the space ${\rm FP}$ of stochastic processes with general filtrations. We also show that $({\rm FP}, \mathcal{AW})$ exhibits several desirable properties. Specifically, it is Polish, martingales form a closed subset and approximation results such as Donsker's theorem extend to $\mathcal{AW}$.

math.PR↗

arXiv2025

FAST drift scan survey for HI intensity mapping: simulation on Bayesian-stacking-based HI mass function estimation

This study investigates the estimation of the neutral hydrogen (HI) mass function (HIMF) using a Bayesian stacking approach with simulated data for the Five-hundred-meter Aperture Spherical radio Telescope (FAST) HI intensity mapping (HIIM) drift-scan surveys. Using data from the IllustrisTNG simulation, we construct HI sky cubes at redshift $z\sim0.1$ and the corresponding optical galaxy catalogs, simulating FAST observations under various survey strategies, including pilot, deep-field, and ultradeep-field surveys. The HIMF is measured for distinct galaxy populations -- classified by optical properties into red, blue, and bluer galaxies -- and injected with systematic effects such as observational noise and flux confusion caused by the FAST beam. The results show that Bayesian stacking significantly enhances HIMF measurements. For red and blue galaxies, the HIMF can be well constrained with pilot surveys, while deeper surveys are required for the bluer galaxy population. Our analysis also reveals that sample variance dominates over observational noise, emphasizing the importance of wide-field surveys to improve constraints. Furthermore, flux confusion shifts the HIMF toward higher masses, which we address using a transfer function for correction. Finally, we explore the effects of intrinsic sample incompleteness and propose a framework to quantify its impact. This work lays the groundwork for future \hiMF studies with FAST HIIM, addressing key challenges and enabling robust analyses of HI content across galaxy populations.

astro-ph.CO↗

arXiv2025

Study on a Fast Solver for Combined Field Integral Equations of 3D Conducting Bodies Based on Graph Neural Networks

In this paper, we present a graph neural networks (GNNs)-based fast solver (GraphSolver) for solving combined field integral equations (CFIEs) of 3D conducting bodies. Rao-Wilton-Glisson (RWG) basis functions are employed to discretely and accurately represent the geometry of 3D conducting bodies. A concise and informative graph representation is then constructed by treating each RWG function as a node in the graph, enabling the flow of current between nodes. With the transformed graphs, GraphSolver is developed to directly predict real and imaginary parts of the x, y and z components of the surface current densities at each node (RWG function). Numerical results demonstrate the efficacy of GraphSolver in solving CFIEs for 3D conducting bodies with varying levels of geometric complexity, including basic 3D targets, missile-shaped targets, and airplane-shaped targets.

cs.LG↗

arXiv2025

Testing the Cosmic Distance Duality Relation Using Strong Gravitational Lensing Time Delays and Type Ia Supernovae

We present a comprehensive test of the cosmic distance duality relation (DDR) using a combination of strong gravitational lensing (SGL) time delay measurements and Type Ia supernovae (SNe Ia) data. We investigate three different parameterizations of potential DDR violations. To bridge the gap between SGL and SNe Ia datasets, we implement an artificial neural network (ANN) approach to reconstruct the distance modulus of SNe Ia. Our analysis uniquely considers both scenarios where the absolute magnitude of SNe Ia ($M_B$) is treated as a free parameter and where it is fixed to a Cepheid-calibrated value. Using a sample of six SGL systems and the Pantheon+ SNe Ia dataset, we find no statistically significant evidence for DDR violations across all parameterizations. The consistency of our findings across different parameterizations not only reinforces confidence in the standard DDR but also demonstrates the robustness of our analytical approach.

astro-ph.CO↗

arXiv2025

Convergence Rate of Particle System for Second-order PDEs On Wasserstein Space

In this paper, we provide a convergence rate for particle approximations of a class of second-order PDEs on Wasserstein space. We show that, up to some error term, the infinite-dimensional inf(sup)-convolution of the finite-dimensional value function yields a super (sub)-viscosity solution to the PDEs on Wasserstein space. Hence, we obtain a convergence rate using a comparison principle of such PDEs on Wasserstein space. Our argument is purely analytic and relies on the regularity of value functions established in \cite{DaJaSe23}.

math.AP↗

arXiv2025

The Backpropagation of the Wave Network

This paper provides an in-depth analysis of Wave Network, a novel token representation method derived from the Wave Network, designed to capture both global and local semantics of input text through wave-inspired complex vectors. In complex vector token representation, each token is represented with a magnitude component, capturing the global semantics of the entire input text, and a phase component, encoding the relationships between individual tokens and the global semantics. Building on prior research that demonstrated the effectiveness of wave-like operations, such as interference and modulation, during forward propagation, this study investigates the convergence behavior, backpropagation characteristics, and embedding independence within the Token2Wave framework. A detailed computational complexity analysis shows that Token2Wave can significantly reduce video memory usage and training time compared to BERT. Gradient comparisons for the [CLS] token, total input text, and classifier parameters further highlight Token2Wave's unique characteristics. This research offers new insights into wave-based token representations, demonstrating their potential to enable efficient and computationally friendly language model architectures.

cs.CL↗

arXiv2025

Superradiant growth anomaly magnification in evolution of vector bosonic condensates bounded by a Kerr black hole with near-horizon reflection

Ultralight vector particles can form evolving condensates around a Kerr black hole (BH) due to superradiant instability. We study the effect of near-horizon reflection on the evolution of this system: by matching three pieces of asymptotic expansions of the Proca equation in Kerr metric and considering the leading order in the electric mode, we present explicit analytical expressions for the corrected spectrum and the superradiant instability rates. Particularly, in high-spin BH cases, we identify an anomalous situation where the superadiance rate is temporarily increased by the reflection parameter $\mathcal{R}$, which also occurs in the scalar scenario, but is largely magnified in vector condensates due to a faster growth rate in dominant mode. We point out the condition for the growth anomaly in the adiabatic case is that information carried per particle exceeds a certain value $δI/δN>2πk_\text{B} \sqrt{(1+\mathcal{R})/(1-\mathcal{R})}$. We further construct several featured quantities to illustrate it, and formalize the anomaly-induced gravitational wave strain deformation.

gr-qc↗

arXiv2025

RSAR: Restricted State Angle Resolver and Rotated SAR Benchmark

Rotated object detection has made significant progress in the optical remote sensing. However, advancements in the Synthetic Aperture Radar (SAR) field are laggard behind, primarily due to the absence of a large-scale dataset. Annotating such a dataset is inefficient and costly. A promising solution is to employ a weakly supervised model (e.g., trained with available horizontal boxes only) to generate pseudo-rotated boxes for reference before manual calibration. Unfortunately, the existing weakly supervised models exhibit limited accuracy in predicting the object's angle. Previous works attempt to enhance angle prediction by using angle resolvers that decouple angles into cosine and sine encodings. In this work, we first reevaluate these resolvers from a unified perspective of dimension mapping and expose that they share the same shortcomings: these methods overlook the unit cycle constraint inherent in these encodings, easily leading to prediction biases. To address this issue, we propose the Unit Cycle Resolver, which incorporates a unit circle constraint loss to improve angle prediction accuracy. Our approach can effectively improve the performance of existing state-of-the-art weakly supervised methods and even surpasses fully supervised models on existing optical benchmarks (i.e., DOTA-v1.0 dataset). With the aid of UCR, we further annotate and introduce RSAR, the largest multi-class rotated SAR object detection dataset to date. Extensive experiments on both RSAR and optical datasets demonstrate that our UCR enhances angle prediction accuracy. Our dataset and code can be found at: https://github.com/zhasion/RSAR.

cs.CV↗

arXiv2025

Exciting games and Monge-Ampère equations

We consider a competition between $d+1$ players, and aim to identify the "most exciting game'' of this kind. This is translated, mathematically, into a stochastic optimization problem over martingales that live on the $d$-dimensional subprobability simplex $Δ$ and terminate on the vertices of $Δ$ (so-called win-martingales), with a cost function related to a scaling limit of Shannon entropies. We uncover a surprising connection between this problem and the seemingly unrelated field of Monge-Ampère equations: If $g$ solves \begin{equation*} \begin{cases} g(x)=\log \det\left(\frac{1}{2}\nabla^2 g(x)\right), \quad \, \ \ \ \ \, \, \, \, x \in Δ, \\ g(x)=\infty, \quad \quad \quad \quad \ \ \ \ \ \quad \quad \, \ \ \ \ \ x\in \partial Δ, \end{cases} \end{equation*} then the winning-probability of the players in the most exciting game is described by $$dM_s=\sqrt{\frac{2 (\nabla^2 g(M_s))^{-1}}{1-s} } \, dB_s.$$ To formalize this, a detailed quantitative analysis of the Monge-Ampère equation for $g$ is crucial. This is then leveraged to prove that $M$ is indeed an optimal win-martingale.

math.PR↗

arXiv2025

Viscosity Solutions of Fully second-order HJB Equations in the Wasserstein Space

In this paper, we show that the value functions of mean field control problems with common noise are the unique viscosity solutions to fully second-order Hamilton-Jacobi-Bellman equations, in a Crandall-Lions-like framework. We allow the second-order derivative in measure to be state-dependent and thus infinite-dimensional, rather than derived from a finite-dimensional operator, hence the term ''fully''. Our argument leverages the construction of smooth approximations from particle systems developed by Cosso, Gozzi, Kharroubi, Pham, and Rosestolato [Trans. Amer. Math. Soc., 2023], and the compactness argument via penalization of measure moments in Soner and Yan [Appl. Math. Optim., 2024]. Our work addresses unbounded dynamics and state-dependent common noise volatility, and to our knowledge, this is the first result of its kind in the literature.

math.OC↗

arXiv2024

Bethe ansatz approach for the steady state of the asymmetric simple exclusion process with open boundaries

We study the asymmetric simple exclusion process with non-diagonal boundary terms under a specific constraint. A symmetric chiral basis is constructed and a special string solution of the Bethe ansatz equations corresponding to the steady state is presented. Using the coordinate Bethe ansatz method, we derive a concise expression for the steady state. The current and density profile in the steady state are also studied.

math-ph↗

arXiv2024

ArgMed-Agents: Explainable Clinical Decision Reasoning with LLM Disscusion via Argumentation Schemes

There are two main barriers to using large language models (LLMs) in clinical reasoning. Firstly, while LLMs exhibit significant promise in Natural Language Processing (NLP) tasks, their performance in complex reasoning and planning falls short of expectations. Secondly, LLMs use uninterpretable methods to make clinical decisions that are fundamentally different from the clinician's cognitive processes. This leads to user distrust. In this paper, we present a multi-agent framework called ArgMed-Agents, which aims to enable LLM-based agents to make explainable clinical decision reasoning through interaction. ArgMed-Agents performs self-argumentation iterations via Argumentation Scheme for Clinical Discussion (a reasoning mechanism for modeling cognitive processes in clinical reasoning), and then constructs the argumentation process as a directed graph representing conflicting relationships. Ultimately, use symbolic solver to identify a series of rational and coherent arguments to support decision. We construct a formal model of ArgMed-Agents and present conjectures for theoretical guarantees. ArgMed-Agents enables LLMs to mimic the process of clinical argumentative reasoning by generating explanations of reasoning in a self-directed manner. The setup experiments show that ArgMed-Agents not only improves accuracy in complex clinical decision reasoning problems compared to other prompt methods, but more importantly, it provides users with decision explanations that increase their confidence.

cs.AI↗

arXiv2024

MMLU-CF: A Contamination-free Multi-task Language Understanding Benchmark

Multiple-choice question (MCQ) datasets like Massive Multitask Language Understanding (MMLU) are widely used to evaluate the commonsense, understanding, and problem-solving abilities of large language models (LLMs). However, the open-source nature of these benchmarks and the broad sources of training data for LLMs have inevitably led to benchmark contamination, resulting in unreliable evaluation results. To alleviate this issue, we propose a contamination-free and more challenging MCQ benchmark called MMLU-CF. This benchmark reassesses LLMs' understanding of world knowledge by averting both unintentional and malicious data leakage. To avoid unintentional data leakage, we source data from a broader domain and design three decontamination rules. To prevent malicious data leakage, we divide the benchmark into validation and test sets with similar difficulty and subject distributions. The test set remains closed-source to ensure reliable results, while the validation set is publicly available to promote transparency and facilitate independent verification. Our evaluation of mainstream LLMs reveals that the powerful GPT-4o achieves merely a 5-shot score of 73.4% and a 0-shot score of 71.9% on the test set, which indicates the effectiveness of our approach in creating a more rigorous and contamination-free evaluation standard. The GitHub repository is available at https://github.com/microsoft/MMLU-CF and the dataset refers to https://huggingface.co/datasets/microsoft/MMLU-CF.

cs.CL↗

arXiv2024

Stochastic linear quadratic optimal control problems with regime-switching jumps in infinite horizon

This paper investigates a stochastic linear-quadratic (SLQ, for short) control problem regulated by a time-invariant Markov chain in infinite horizon. Under the $L^2$-stability framework, we study a class of linear backward stochastic differential equations (BSDE, for short) in infinite horizon and discuss the open-loop and closed-loop solvabilities of the SLQ problem. The open-loop solvability is characterized by the solvability of a system of coupled forward-backward stochastic differential equations (FBSDEs, for short) in infinite horizon and the convexity of the cost functional, and the closed-loop solvability is shown to be equivalent to the open-loop solvability, which in turn is equivalent to the existence of a static stabilizing solution to the associated constrained coupled algebra Riccati equations (CAREs, for short). Under the uniform convexity assumption, we obtain the unique solvability of associated CAREs and construct the corresponding closed-loop optimal strategy. Finally, we also solve a class of discounted SLQ problems and give two concrete examples to illustrate the results developed in the earlier sections.

math.OC↗

arXiv2024

Scatter correction for photon-counting detector based CBCT imaging

Objective: The aim of this study is to validate the effectiveness of an energy-modulated scatter correction method in suppressing scatter in photon-counting detector (PCD)-based cone beam CT (CBCT) imaging. Approach: The scatter correction method, named e-Grid, which was initially applied to dual-layer flat-panel detector (DLFPD)-based CBCT imaging, was tested for its performance in PCD-CBCT imaging. Benchtop PCD-CBCT imaging experiments were conducted to verify the effectiveness of the e-Grid method. Additionally, quantitative metrics were measured from these experimental results. Main results: It was found that the use of the e-Grid method could significantly eliminate cupping artifacts caused by Compton scatter in PCD-CBCT imaging. Meanwhile, its effectiveness was observed in both low- and high-energy images, as well as for objects of varying sizes. Quantitative results showed that the e-Grid method could reduce scatter artifacts by at least 71% in low-energy images and 75% in high-energy images. Significance: It was demonstrated that the scatter correction method originally applied to DLFPD-based CBCT could also perform well in PCD-CBCT, showing that the e-Grid method has great potential for application in other spectral CBCT imaging systems.

physics.med-ph↗