Searcharxiv⌕ Search

arXiv subjects

Zexin Wang

Publications and source records attributed to Zexin Wang.

28 records · Page 2Linked to original sources
arXiv2025

Flow-of-Action: SOP Enhanced LLM-Based Multi-Agent System for Root Cause Analysis

In the realm of microservices architecture, the occurrence of frequent incidents necessitates the employment of Root Cause Analysis (RCA) for swift issue resolution. It is common that a serious incident can take several domain experts hours to identify the root cause. Consequently, a contemporary trend involves harnessing Large Language Models (LLMs) as automated agents for RCA. Though the recent ReAct framework aligns well with the Site Reliability Engineers (SREs) for its thought-action-observation paradigm, its hallucinations often lead to irrelevant actions and directly affect subsequent results. Additionally, the complex and variable clues of the incident can overwhelm the model one step further. To confront these challenges, we propose Flow-of-Action, a pioneering Standard Operation Procedure (SOP) enhanced LLM-based multi-agent system. By explicitly summarizing the diagnosis steps of SREs, SOP imposes constraints on LLMs at crucial junctures, guiding the RCA process towards the correct trajectory. To facilitate the rational and effective utilization of SOPs, we design an SOP-centric framework called SOP flow. SOP flow contains a series of tools, including one for finding relevant SOPs for incidents, another for automatically generating SOPs for incidents without relevant ones, and a tool for converting SOPs into code. This significantly alleviates the hallucination issues of ReAct in RCA tasks. We also design multiple auxiliary agents to assist the main agent by removing useless noise, narrowing the search space, and informing the main agent whether the RCA procedure can stop. Compared to the ReAct method's 35.50% accuracy, our Flow-of-Action method achieves 64.01%, meeting the accuracy requirements for RCA in real-world systems.

cs.SE↗
arXiv2025

The resolutions of generalized co-letterplace ideals and their powers

We present a natural and explicit multigraded minimal free resolution for each generalized co-letterplace ideal (see Definition 1.1). Our resolution differs significantly from the ones presented in the works of Ene et al. \cite{EHM} and D'Al{`ı} et al. \cite{DFN}. Additionally, we show that each power of a large class of generalized co-letterplace ideals can be represented as the quotient of another generalized co-letterplace ideal by a regular sequence of variable differences. Finally, we provide a new class of simplicial spheres

math.AC↗
arXiv2024

Ferroelectric Materials for Synaptic Transistors and Their Neuromorphic Applications

After more than a hundred years of development, ferroelectric materials have demonstrated their strong potential to people, and more and more ferroelectric materials are being used in the research of ferroelectric transistors (FeFETs). As a new generation of neuromorphic devices, ferroelectric materials have attracted people's attention due to their powerful functions and many characteristics. This article summarizes the development of ferroelectric material systems in recent years and discusses the simulation of artificial synapses. The mainstream ferroelectric materials are divided into traditional perovskite structure, fluorite structure, organic polymer, and new 2D van der Waals ferroelectricity. The principles, research progress, and optimization for brain like computers of each material system are introduced, and the latest application progress is summarized. Finally, the scope of application of different material systems is discussed, with the aim of helping people screen out different material systems based on different needs.

physics.app-ph↗
arXiv2024

Multiplicity of powers of path ideals of a line graph

Let $S=K[x_1,\ldots,x_n]$ and let $I$ be the $t$-path ideal of the line graph $L_n$ with $n$-vertices. It is shown that the set of associated prime ideals of $I^s$ is equal to the set of minimal prime ideals of $I$ for all $s\geq 1$, and we provide an explicit description of these prime ideals. Additionally, as the main contribution of this paper, we derive an explicit formula for the multiplicity of $S/I^s$ for all $s\geq 1$, revealing that it is a polynomial in $s$ from the beginning.

math.AC↗
arXiv2024

The edge rings of compact graphs

We define a simple graph as compact if it lacks even cycles and satisfies the odd-cycle condition. Our focus is on classifying all compact graphs and examining the characteristics of their edge rings. Let $G$ be a compact graph and $\mathbb{K}[G]$ be its edge ring. Specifically, we demonstrate that the Cohen-Macaulay type and the projective dimension of $\mathbb{K}[G]$ are both equal to the number of induced cycles of $G$ minus one, and that the regularity of $\mathbb{K}[G]$ is equal to the matching number of $G_0$. Here, $G_0$ is obtained from $G$ by removing the vertices of degree one successively, resulting in a graph where every vertex has a degree greater than 1.

math.AC↗
arXiv2024

Revisiting VAE for Unsupervised Time Series Anomaly Detection: A Frequency Perspective

Time series Anomaly Detection (AD) plays a crucial role for web systems. Various web systems rely on time series data to monitor and identify anomalies in real time, as well as to initiate diagnosis and remediation procedures. Variational Autoencoders (VAEs) have gained popularity in recent decades due to their superior de-noising capabilities, which are useful for anomaly detection. However, our study reveals that VAE-based methods face challenges in capturing long-periodic heterogeneous patterns and detailed short-periodic trends simultaneously. To address these challenges, we propose Frequency-enhanced Conditional Variational Autoencoder (FCVAE), a novel unsupervised AD method for univariate time series. To ensure an accurate AD, FCVAE exploits an innovative approach to concurrently integrate both the global and local frequency features into the condition of Conditional Variational Autoencoder (CVAE) to significantly increase the accuracy of reconstructing the normal data. Together with a carefully designed "target attention" mechanism, our approach allows the model to pick the most useful information from the frequency domain for better short-periodic trend construction. Our FCVAE has been evaluated on public datasets and a large-scale cloud system, and the results demonstrate that it outperforms state-of-the-art methods. This confirms the practical applicability of our approach in addressing the limitations of current VAE-based anomaly detection models.

cs.LG↗
arXiv2023

On powers of the cover ideals of graphs

For a simple graph $G$, assume that $J(G)$ is the vertex cover ideal of $G$ and $J(G)^{(s)}$ is the $s$-th symbolic power of $J(G)$. We prove that $(J(C)^{(s)})=(J(C)^s)$ for all $s\geq 1$ and for all odd cycle $C$. For a simplicial complex $Δ$, we show that $Δ$ is vertex decomposable if $I_Δ^{\vee}$ is weakly polymatroidal. Let $W=G^π$ be a fully clique-whiskering graph, we prove that $J(W)^s$ is weakly polymatroidal for all $s\geq 1$.

math.AC↗
arXiv2022

Tight Regret Bounds for Noisy Optimization of a Brownian Motion

We consider the problem of Bayesian optimization of a one-dimensional Brownian motion in which the $T$ adaptively chosen observations are corrupted by Gaussian noise. We show that as the smallest possible expected cumulative regret and the smallest possible expected simple regret scale as $Ω(σ\sqrt{T / \log (T)}) \cap \mathcal{O}(σ\sqrt{T} \cdot \log T)$ and $Ω(σ/ \sqrt{T \log (T)}) \cap \mathcal{O}(σ\log T / \sqrt{T})$ respectively, where $σ^2$ is the noise variance. Thus, our upper and lower bounds are tight up to a factor of $\mathcal{O}( (\log T)^{1.5} )$. The upper bound uses an algorithm based on confidence bounds and the Markov property of Brownian motion (among other useful properties), and the lower bound is based on a reduction to binary hypothesis testing.

cs.LG↗
arXiv2018

Parameter-Free Spatial Attention Network for Person Re-Identification

Global average pooling (GAP) allows to localize discriminative information for recognition [40]. While GAP helps the convolution neural network to attend to the most discriminative features of an object, it may suffer if that information is missing e.g. due to camera viewpoint changes. To circumvent this issue, we argue that it is advantageous to attend to the global configuration of the object by modeling spatial relations among high-level features. We propose a novel architecture for Person Re-Identification, based on a novel parameter-free spatial attention layer introducing spatial relations among the feature map activations back to the model. Our spatial attention layer consistently improves the performance over the model without it. Results on four benchmarks demonstrate a superiority of our model over the state-of-the-art achieving rank-1 accuracy of 94.7% on Market-1501, 89.0% on DukeMTMC-ReID, 74.9% on CUHK03-labeled and 69.7% on CUHK03-detected.

cs.CV↗