arXiv subjects

Yichen Wang

Publications and source records attributed to Yichen Wang.

At least 55 records · Page 3Linked to original sources

arXiv2026

Channel-Aware Conditional Diffusion Model for Secure MU-MISO Communications

While information securityis a fundamental requirement for wireless communications, conventional optimization based approaches often struggle with real-time implementation, and deep models, typically discriminative in nature, may lack the ability to cope with unforeseen scenarios. To address this challenge, this paper investigates the design of legitimate beamforming and artificial noise (AN) to achieve physical layer security by exploiting the conditional diffusion model. Specifically, we reformulate the security optimization as a conditional generative process, using a diffusion model to learn the inherent distribution of near-optimal joint beamforming and AN strategies. We employ a U-Net architecture with cross-attention to integrate channel state information, as the basis for the generative process. Moreover, we fine-tune the trained model using an objective incorporating the sum secrecy rate such that the security performance is further enhanced. Finally, simulation results validate the learning process convergence and demonstrate that the proposed generative method achieves superior secrecy performance across various scenarios as compared with the baselines.

eess.SP↗

arXiv2026

Smell with Genji: Rediscovering Human Perception through an Olfactory Game with AI

Olfaction plays an important role in human perception, yet its subjective and ephemeral nature makes it difficult to articulate, compare, and share across individuals. Traditional practices like the Japanese incense game Genji-ko offer one way to structure olfactory experience through shared interpretation. In this work, we present Smell with Genji, an AI-mediated olfactory interaction system that reinterprets Genji-ko as a collaborative human-AI sensory experience. By integrating a game setup, a mobile application, and an LLM-powered co-smelling partner equipped with olfactory sensing and LLM-based conversation, the system invites participants to compare scents and construct Genji-mon patterns, fostering reflection through a dialogue that highlights the alignment and discrepancies between human and machine perception. This work illustrates how sensing-enabled AI can participate in olfactory experience alongside users, pointing toward new possibilities for AI-supported sensory interaction and reflection in HCI.

cs.HC↗

arXiv2026

The Turán number of Berge matchings

Given a graph $F$, an $r$-uniform hypergraph $\mathcal{H}$ is a {\em Berge-$F$} if there is a bijection $ϕ:E(F)\to E(\mathcal{H})$ such that $e\subseteq ϕ(e)$ for each $e\in E(F)$. Given a family $\mathcal{F}$ of $r$-uniform hypergraphs, an $r$-uniform hypergraph is $\mathcal{F}$-free if it does not contain any member of $\mathcal{F}$ as a subhypergraph. The Turán number of $\mathcal{F}$ is the maximum number of hyperedges in an $\mathcal{F}$-free $r$-graph on $n$ vertices. Let $M_{s+1}$ denote a matching of size $s+1$, i.e., the graph consisting of $s+1$ independent edges. Khormali and Palmer [\textit{European J. Combin.} 102 (2022) 103506] completely determined the Turán number of Berge matchings for sufficiently large $n$. Subsequently, Kang, Ni, and Shan [\textit{Discrete Math.} 345 (2022) 112901] determined the exact value of the Turán number of Berge-$M_{s+1}$ for all $n$ when $r \le s-1$ or $r \ge 2s+2$. In this paper, we settle the final open case $s \le r \le 2s+1$, thereby completing the determination of the Turán number of Berge matchings.

math.CO↗

arXiv2026

The maximum number of triangles in graphs without the square of a path

The generalized Turán number for $H$ of $G$, denoted by $\ex(n,H,G)$, is the maximum number of copies of $H$ in an $n$-vertex $G$-free graph. When $H$ is an edge, $\ex(n,H,G)$ is the classical Turán number $\ex(n,G)$. Let $P_k$ be the path with $k$ vertices. The square of $P_k$, denoted by $P_k^2$, is obtained by joining the pairs of vertices with distance at most two in $P_k$. The Turán number of $P_k^2$, $\ex(n, P_k^2)$, was determined by several researchers. When $k=3$, $P_3^2$ is the triangle and $\ex(n, P_3^2)$ is well-known from Mantel's theorem. When $k=4$, $\ex(n, P_4^2)$ was solved by Dirac in a more general context. When $k=5,6$, the problem was solved by Xiao, Katona, Xiao, and Zamora. For general $k \ge 7$, the problem was solved by Yuan in a more general context. Recently, Mukherjee determined the generalized Turán number $\ex(n, K_3, P_5^2)$. In this paper, we determine the exact value of $\ex(n, K_3, P_6^2)$ and characterize all the extremal graphs for $n \ge 11$.

math.CO↗

arXiv2026

Towards Efficient 3D Object Detection for Vehicle-Infrastructure Collaboration via Risk-Intent Selection

Vehicle-Infrastructure Collaborative Perception (VICP) is pivotal for resolving occlusion in autonomous driving, yet the trade-off between communication bandwidth and feature redundancy remains a critical bottleneck. While intermediate fusion mitigates data volume compared to raw sharing, existing frameworks typically rely on spatial compression or static confidence maps, which inefficiently transmit spatially redundant features from non-critical background regions. To address this, we propose Risk-intent Selective detection (RiSe), an interaction-aware framework that shifts the paradigm from identifying visible regions to prioritizing risk-critical ones. Specifically, we introduce a Potential Field-Trajectory Correlation Model (PTCM) grounded in potential field theory to quantitatively assess kinematic risks. Complementing this, an Intention-Driven Area Prediction Module (IDAPM) leverages ego-motion priors to proactively predict and filter key Bird's-Eye-View (BEV) areas essential for decision-making. By integrating these components, RiSe implements a semantic-selective fusion scheme that transmits high-fidelity features only from high-interaction regions, effectively acting as a feature denoiser. Extensive experiments on the DeepAccident dataset demonstrate that our method reduces communication volume to 0.71\% of full feature sharing while maintaining state-of-the-art detection accuracy, establishing a competitive Pareto frontier between bandwidth efficiency and perception performance.

cs.CV↗

arXiv2025

Early GVHD Prediction in Liver Transplantation via Multi-Modal Deep Learning on Imbalanced EHR Data

Graft-versus-host disease (GVHD) is a rare but often fatal complication in liver transplantation, with a very high mortality rate. By harnessing multi-modal deep learning methods to integrate heterogeneous and imbalanced electronic health records (EHR), we aim to advance early prediction of GVHD, paving the way for timely intervention and improved patient outcomes. In this study, we analyzed pre-transplant electronic health records (EHR) spanning the period before surgery for 2,100 liver transplantation patients, including 42 cases of graft-versus-host disease (GVHD), from a cohort treated at Mayo Clinic between 1992 and 2025. The dataset comprised four major modalities: patient demographics, laboratory tests, diagnoses, and medications. We developed a multi-modal deep learning framework that dynamically fuses these modalities, handles irregular records with missing values, and addresses extreme class imbalance through AUC-based optimization. The developed framework outperforms all single-modal and multi-modal machine learning baselines, achieving an AUC of 0.836, an AUPRC of 0.157, a recall of 0.768, and a specificity of 0.803. It also demonstrates the effectiveness of our approach in capturing complementary information from different modalities, leading to improved performance. Our multi-modal deep learning framework substantially improves existing approaches for early GVHD prediction. By effectively addressing the challenges of heterogeneity and extreme class imbalance in real-world EHR, it achieves accurate early prediction. Our proposed multi-modal deep learning method demonstrates promising results for early prediction of a GVHD in liver transplantation, despite the challenge of extremely imbalanced EHR data.

cs.LG↗

arXiv2025

Multiscale spatiotemporal heterogeneity analysis of bike-sharing system's self-loop phenomenon: Evidence from Shanghai

Bike-sharing is an environmentally friendly shared mobility mode, but its self-loop phenomenon, where bikes are returned to the same station after several time usage, significantly impacts equity in accessing its services. Therefore, this study conducts a multiscale analysis with a spatial autoregressive model and double machine learning framework to assess socioeconomic features and geospatial location's impact on the self-loop phenomenon at metro stations and street scales. The results reveal that bike-sharing self-loop intensity exhibits significant spatial lag effect at street scale and is positively associated with residential land use. Marginal treatment effects of residential land use is higher on streets with middle-aged residents, high fixed employment, and low car ownership. The multimodal public transit condition reveals significant positive marginal treatment effects at both scales. To enhance bike-sharing cooperation, we advocate augmenting bicycle availability in areas with high metro usage and low bus coverage, alongside implementing adaptable redistribution strategies.

cs.LG↗

arXiv2025

Linear recoloring diameter of degenerate chordal graphs and bounded treewidth graphs

Let $G$ be a graph on $n$ vertices and $t$ an integer. The reconfiguration graph of $G$, denoted by $R_t(G)$, consists of all $t$-colorings of $G$ and two $t$-colorings are adjacent if they differ on exactly one vertex. The $t$-recoloring diameter of $G$ is the diameter of $R_t(G)$. For a $d$-degenerate graph $G$, $R_t(G)$ is connected when $t \ge d+2$~(Dyer et al., 2006). Furthermore, the $t$-recoloring diameter is $O(n^2)$ when $t \ge 3(d+1)/2$~(Bousquet et al., 2022), and it is $O(n)$ when $t \ge 2d+2$~(Bousquet and Perarnau, 2016). For a $d$-degenerate and chordal graph $G$, the $t$-recoloring diameter of $G$ is $O(n^2)$ when $t \ge d+2$~(Bonamy et al. 2014). If $G$ is a graph of treewidth at most $k$, then $G$ is also $k$-degenerate, and the previous results hold. Moreover, when $t \ge k+2$, the $t$-recoloring diameter is $O(n^2)$~(Bonamy and Bousquet, 2013). When $k=2$, the $t$-recoloring diameter of $G$ is linear when $t \ge 5$~(Bartier, Bousquet and Heinrich, 2021) and the result is tight. In this paper, we prove that if $G$ is $d$-degenerate and chordal, then the $t$-recoloring diameter of $G$ is $O(n)$ when $t \ge 2d+1$. Moreover, if the treewidth of $G$ is at most $k$, then the $t$-recoloring diameter is $O(n)$ when $t \ge 2k+1$. This result is a generalization of the previous results on graphs of treewidth at most two.

math.CO↗

arXiv2025

The $φ$ Curve: The Shape of Generalization through the Lens of Norm-based Capacity Control

Understanding how the test risk scales with model complexity is a central question in machine learning. Classical theory is challenged by the learning curves observed for large over-parametrized deep networks. Capacity measures based on parameter count typically fail to account for these empirical observations. To tackle this challenge, we consider norm-based capacity measures and develop our study for random features based estimators, widely used as simplified theoretical models for more complex networks. In this context, we provide a precise characterization of how the estimator's norm concentrates and how it governs the associated test error. Our results show that the predicted learning curve admits a phase transition from under- to over-parameterization, but no double descent behavior. This confirms that more classical U-shaped behavior is recovered considering appropriate capacity measures based on models norms rather than size. From a technical point of view, we leverage deterministic equivalence as the key tool and further develop new deterministic quantities which are of independent interest.

stat.ML↗

arXiv2025

Edge version of the inducibility via the entropy method

The inducibility of a graph $H$ is about the maximum number of induced copies of $H$ in a graph on $n$ vertices. We consider its edge version, that is, the maximum number of induced copies of $H$ in a graph with $m$ edges. Let $c(G,H)$ be the number of induced copies of $H$ in $G$ and $ρ(H,m) = \max \{c(G,H) \mid |E(G)| = m\}$. For any graph $H$, we prove that $ρ(H,m) = Θ(m^{α_f(H)})$ where $α_f(H)$ is the fractional independence number of $H$. Therefore, we now focus on the constant factor in front of $m^{α_f(H)}$. In this paper, we give some results of $ρ(H,m)$ when $H$ is a cycle or path. We conjecture that for any cycle $C_k$ with $k \ge 5$, $ρ(C_k,m)= (1+o(1))\left( m/k\right)^{k/2}$ and the bound achieves by the blow up of $C_k$. For even cycles, we establish an upper bound with an extra constant factor. For odd cycles, we can only establish an upper bound with an extra factor depending on $k$. We prove that $ρ(P_{2l},m) \le \frac{m^l}{2(l-1)^{l-1}}$ and $ρ(P_{2l+1},m) \le \frac{m^{l+1}}{4l^l}$, where $l \ge 2$. We also conjecture the asymptotic value of $ρ(P_k, m)$. The entropy method is mainly used to prove our results.

math.CO↗

arXiv2025

Unraveling Misinformation Propagation in LLM Reasoning

Large Language Models (LLMs) have demonstrated impressive capabilities in reasoning, positioning them as promising tools for supporting human problem-solving. However, what happens when their performance is affected by misinformation, i.e., incorrect inputs introduced by users due to oversights or gaps in knowledge? Such misinformation is prevalent in real-world interactions with LLMs, yet how it propagates within LLMs' reasoning process remains underexplored. Focusing on mathematical reasoning, we present a comprehensive analysis of how misinformation affects intermediate reasoning steps and final answers. We also examine how effectively LLMs can correct misinformation when explicitly instructed to do so. Even with explicit instructions, LLMs succeed less than half the time in rectifying misinformation, despite possessing correct internal knowledge, leading to significant accuracy drops (10.02% - 72.20%), and the degradation holds with thinking models (4.30% - 19.97%). Further analysis shows that applying factual corrections early in the reasoning process most effectively reduces misinformation propagation, and fine-tuning on synthesized data with early-stage corrections significantly improves reasoning factuality. Our work offers a practical approach to mitigating misinformation propagation.

cs.CL↗

arXiv2025

Treewidth of generalized Hamming graph, bipartite Kneser graph and generalized Petersen graph

Let $t,q$ and $n$ be positive integers. Write $[q] = \{1,2,\ldots,q\}$. The generalized Hamming graph $H(t,q,n)$ is the graph whose vertex set is the cartesian product of $n$ copies of $[q]$ ($q\ge 2$), where two vertices are adjacent if their Hamming distance is at most $t$. In particular, $H(1,q,n)$ is the well-known Hamming graph and $H(1,2,n)$ is the hypercube. In 2006, Chandran and Kavitha described the asymptotic value of $tw(H(1,q,n))$, where $tw(G)$ denotes the treewidth of $G$. In this paper, we give the exact pathwidth of $H(t,2,n)$ and show that $tw(H(t,q,n)) = Θ(tq^n/\sqrt{n})$ when $n$ goes to infinity. Based on those results, we show that the treewidth of the bipartite Kneser graph $BK(n,k)$ is $\binom{n}{k} - 1$ when $n$ is sufficiently large relative to $k$ and the bounds of $tw(BK(2k+1,k))$ are given. Moreover, we present the bounds of the treewidth of the generalized Petersen graph.

math.CO↗

arXiv2025

ADVEDM:Fine-grained Adversarial Attack against VLM-based Embodied Agents

Vision-Language Models (VLMs), with their strong reasoning and planning capabilities, are widely used in embodied decision-making (EDM) tasks in embodied agents, such as autonomous driving and robotic manipulation. Recent research has increasingly explored adversarial attacks on VLMs to reveal their vulnerabilities. However, these attacks either rely on overly strong assumptions, requiring full knowledge of the victim VLM, which is impractical for attacking VLM-based agents, or exhibit limited effectiveness. The latter stems from disrupting most semantic information in the image, which leads to a misalignment between the perception and the task context defined by system prompts. This inconsistency interrupts the VLM's reasoning process, resulting in invalid outputs that fail to affect interactions in the physical world. To this end, we propose a fine-grained adversarial attack framework, ADVEDM, which modifies the VLM's perception of only a few key objects while preserving the semantics of the remaining regions. This attack effectively reduces conflicts with the task context, making VLMs output valid but incorrect decisions and affecting the actions of agents, thus posing a more substantial safety threat in the physical world. We design two variants of based on this framework, ADVEDM-R and ADVEDM-A, which respectively remove the semantics of a specific object from the image and add the semantics of a new object into the image. The experimental results in both general scenarios and EDM tasks demonstrate fine-grained control and excellent attack performance.

cs.CV↗

arXiv2025

Counting induced subgraphs with given intersection sizes

Let $F$ be a graph of order $r$. In this paper, we study the maximum number of induced copies of $F$ with restricted intersections, which highlights the motivation from extremal set theory. Let $L=\{\ell_1,\dots,\ell_s\}\subseteq[0,r-1]$ be an integer set with $s\not\in\{1,r\}$. Let $Ψ_r(n,F,L)$ be the maximum number of induced copies of $F$ in an $n$-vertex graph, where the induced copies of $F$ are $L$-intersecting as a family of $r$-subsets, i.e., for any two induced copies of $F$, the size of their intersection is in $L$. Helliar and Liu initiated a study of the function $Ψ_r(n,K_r,L)$. Very recently, Zhao and Zhang improved their result and showed that $Ψ_r(n,K_r,L)=Θ_{r,L}(n^{s})$ if and only if $\ell_1,\dots,\ell_s,r$ form an arithmetic progression. In this paper, we show that $Ψ_r(n,F,L)=o_{r,L}(n^{s})$ when $\ell_1,\dots,\ell_s,r$ do not form an arithmetic progression. We study the asymptotical result of $Ψ_r(n,C_r,L)$, and determined the asymptotically optimal result when $\ell_1,\dots,\ell_s,r$ form an arithmetic progression and take certain values. We also study the generalized Turán problem, determining the maximum number of $H$, where the copies of $H$ are $L$-intersecting as a family of $r$-subsets. The entropy method is used to prove our results.

math.CO↗

arXiv2025

Agentic Satellite-Augmented Low-Altitude Economy and Terrestrial Networks: A Survey on Generative Approaches

The development of satellite-augmented low-altitude economy and terrestrial networks (SLAETNs) demands intelligent and autonomous systems that can operate reliably across heterogeneous, dynamic, and mission-critical environments. To address these challenges, this survey focuses on enabling agentic artificial intelligence (AI), that is, artificial agents capable of perceiving, reasoning, and acting, through generative AI (GAI) and large language models (LLMs). We begin by introducing the architecture and characteristics of SLAETNs, and analyzing the challenges that arise in integrating satellite, aerial, and terrestrial components. Then, we present a model-driven foundation by systematically reviewing five major categories of generative models: variational autoencoders (VAEs), generative adversarial networks (GANs), generative diffusion models (GDMs), transformer-based models (TBMs), and LLMs. Moreover, we provide a comparative analysis to highlight their generative mechanisms, capabilities, and deployment trade-offs within SLAETNs. Building on this foundation, we examine how these models empower agentic functions across three domains: communication enhancement, security and privacy protection, and intelligent satellite tasks. Finally, we outline key future directions for building scalable, adaptive, and trustworthy generative agents in SLAETNs. This survey aims to provide a unified understanding and actionable reference for advancing agentic AI in next-generation integrated networks.

cs.NI↗

arXiv2025

DPNO: A Dual Path Architecture For Neural Operator

Neural operators have emerged as a powerful tool for solving partial differential equations (PDEs) and other complex scientific computing tasks. However, the performance of single operator block is often limited, thus often requiring composition of basic operator blocks to achieve better per-formance. The traditional way of composition is staking those blocks like feedforward neural networks, which may not be very economic considering parameter-efficiency tradeoff. In this pa-per, we propose a novel dual path architecture that significantly enhances the capabilities of basic neural operators. The basic operator block is organized in parallel two paths which are similar with ResNet and DenseNet. By introducing this parallel processing mechanism, our architecture shows a more powerful feature extraction and solution approximation ability compared with the original model. We demonstrate the effectiveness of our approach through extensive numerical experi-ments on a variety of PDE problems, including the Burgers' equation, Darcy Flow Equation and the 2d Navier-Stokes equation. The experimental results indicate that on certain standard test cas-es, our model achieves a relative improvement of over 30% compared to the basic model. We also apply this structure on two standard neural operators (DeepONet and FNO) selected from different paradigms, which suggests that the proposed architecture has excellent versatility and offering a promising direction for neural operator structure design.

math.NA↗

arXiv2025

Can A Society of Generative Agents Simulate Human Behavior and Inform Public Health Policy? A Case Study on Vaccine Hesitancy

Can we simulate a sandbox society with generative agents to model human behavior, thereby reducing the over-reliance on real human trials for assessing public policies? In this work, we investigate the feasibility of simulating health-related decision-making, using vaccine hesitancy, defined as the delay in acceptance or refusal of vaccines despite the availability of vaccination services (MacDonald, 2015), as a case study. To this end, we introduce the VacSim framework with 100 generative agents powered by Large Language Models (LLMs). VacSim simulates vaccine policy outcomes with the following steps: 1) instantiate a population of agents with demographics based on census data; 2) connect the agents via a social network and model vaccine attitudes as a function of social dynamics and disease-related information; 3) design and evaluate various public health interventions aimed at mitigating vaccine hesitancy. To align with real-world results, we also introduce simulation warmup and attitude modulation to adjust agents' attitudes. We propose a series of evaluations to assess the reliability of various LLM simulations. Experiments indicate that models like Llama and Qwen can simulate aspects of human behavior but also highlight real-world alignment challenges, such as inconsistent responses with demographic profiles. This early exploration of LLM-driven simulations is not meant to serve as definitive policy guidance; instead, it serves as a call for action to examine social simulation for policy development.

cs.MA↗

arXiv2025

Optimal Differentially Private Ranking from Pairwise Comparisons

Data privacy is a central concern in many applications involving ranking from incomplete and noisy pairwise comparisons, such as recommendation systems, educational assessments, and opinion surveys on sensitive topics. In this work, we propose differentially private algorithms for ranking based on pairwise comparisons. Specifically, we develop and analyze ranking methods under two privacy notions: edge differential privacy, which protects the confidentiality of individual comparison outcomes, and individual differential privacy, which safeguards potentially many comparisons contributed by a single individual. Our algorithms--including a perturbed maximum likelihood estimator and a noisy count-based method--are shown to achieve minimax optimal rates of convergence under the respective privacy constraints. We further demonstrate the practical effectiveness of our methods through experiments on both simulated and real-world data.

math.ST↗