arXiv subjects

Yufan Zheng

Publications and source records attributed to Yufan Zheng.

11 recordsLinked to original sources

arXiv2026

Breadth Beats Depth: Improving GCG-Based Jailbreak Optimization with Breadth-Oriented Suffix Search

Optimization-based jailbreak attacks such as Greedy Coordinate Gradient (GCG) achieve strong effectiveness and transferability by optimizing adversarial suffixes on white-box source models. However, existing GCG-based methods rely on averaged adversarial loss and deep greedy search, which can over-emphasize easy-to-jailbreak behaviors and overlook promising regions of the suffix space. We propose BOSS, a plug-and-play framework that improves GCG-based jailbreak optimization through breadth-oriented suffix search. BOSS uses Tail-Focused Adversarial Loss (TFAL), standard source loss, and behavior coverage to select terminal suffixes, then explores multiple short trajectories and selectively continues promising suffixes. Experiments on public benchmarks show that BOSS improves attack success rates across multiple GCG-based methods while reducing optimization time.

cs.CL↗

arXiv2025

Epidemiology-informed Graph Neural Network for Heterogeneity-aware Epidemic Forecasting

Among various spatio-temporal prediction tasks, epidemic forecasting plays a critical role in public health management. Recent studies have demonstrated the strong potential of spatio-temporal graph neural networks (STGNNs) in extracting heterogeneous spatio-temporal patterns for epidemic forecasting. However, most of these methods bear an over-simplified assumption that two locations (e.g., cities) with similar observed features in previous time steps will develop similar infection numbers in the future. In fact, for any epidemic disease, there exists strong heterogeneity of its intrinsic evolution mechanisms across geolocation and time, which can eventually lead to diverged infection numbers in two ``similar'' locations. However, such mechanistic heterogeneity is non-trivial to be captured due to the existence of numerous influencing factors like medical resource accessibility, virus mutations, mobility patterns, etc., most of which are spatio-temporal yet unreachable or even unobservable. To address this challenge, we propose a Heterogeneous Epidemic-Aware Transmission Graph Neural Network (HeatGNN), a novel epidemic forecasting framework. By binding the epidemiology mechanistic model into a GNN, HeatGNN learns epidemiology-informed location embeddings of different locations that reflect their own transmission mechanisms over time. With the time-varying mechanistic affinity graphs computed with the epidemiology-informed location embeddings, a heterogeneous transmission graph network is designed to encode the mechanistic heterogeneity among locations, providing additional predictive signals to facilitate accurate forecasting. Experiments on three benchmark datasets have revealed that HeatGNN outperforms various strong baselines. Moreover, our efficiency analysis verifies the real-world practicality of HeatGNN on datasets of different sizes.

cs.LG↗

arXiv2025

(Sub)Exponential Quantum Speedup for Optimization

We demonstrate provable (sub)exponential quantum speedups in both discrete and continuous optimization, achieved through simple and natural quantum optimization algorithms, namely the quantum adiabatic algorithm for discrete optimization and quantum Hamiltonian descent for continuous optimization. Our result builds on the Gilyén--Hastings--Vazirani (sub)exponential oracle separation for adiabatic quantum computing. With a sequence of perturbative reductions, we compile their construction into two standalone objective functions, whose oracles can be directly leveraged by the plain adiabatic evolution and Schrödinger operator evolution for discrete and continuous optimization, respectively.

quant-ph↗

arXiv2025

Quantum Hamiltonian Descent for Non-smooth Optimization

Non-smooth optimization models play a fundamental role in various disciplines, including engineering, science, management, and finance. However, classical algorithms for solving such models often struggle with convergence speed, scalability, and parameter tuning, particularly in high-dimensional and non-convex settings. In this paper, we explore how quantum mechanics can be leveraged to overcome these limitations. Specifically, we investigate the theoretical properties of the Quantum Hamiltonian Descent (QHD) algorithm for non-smooth optimization in both continuous and discrete time. First, we propose continuous-time variants of the general QHD algorithm and establish their global convergence and convergence rate for non-smooth convex and strongly convex problems through a novel Lyapunov function design. Furthermore, we prove the finite-time global convergence of continuous-time QHD for non-smooth non-convex problems under mild conditions (i.e., locally Lipschitz). In addition, we propose discrete-time QHD, a fully digitized implementation of QHD via operator splitting (i.e., product formula). We find that discrete-time QHD exhibits similar convergence properties even with large time steps. Finally, numerical experiments validate our theoretical findings and demonstrate the computational advantages of QHD over classical non-smooth non-convex optimization algorithms.

math.OC↗

arXiv2024

On the Computational Complexity of Schrödinger Operators

We study computational problems related to the Schrödinger operator $H = -Δ+ V$ in the real space under the condition that (i) the potential function $V$ is smooth and has its value and derivative bounded within some polynomial of $n$ and (ii) $V$ only consists of $O(1)$-body interactions. We prove that (i) simulating the dynamics generated by the Schrödinger operator implements universal quantum computation, i.e., it is BQP-hard, and (ii) estimating the ground energy of the Schrödinger operator is as hard as estimating that of local Hamiltonians with no sign problem (a.k.a. stoquastic Hamiltonians), i.e., it is StoqMA-complete. This result is particularly intriguing because the ground energy problem for general bosonic Hamiltonians is known to be QMA-hard and it is widely believed that $\texttt{StoqMA}\varsubsetneq \texttt{QMA}$.

quant-ph↗

arXiv2023

A quantum-classical performance separation in nonconvex optimization

In this paper, we identify a family of nonconvex continuous optimization instances, each $d$-dimensional instance with $2^d$ local minima, to demonstrate a quantum-classical performance separation. Specifically, we prove that the recently proposed Quantum Hamiltonian Descent (QHD) algorithm [Leng et al., arXiv:2303.01471] is able to solve any $d$-dimensional instance from this family using $\widetilde{\mathcal{O}}(d^3)$ quantum queries to the function value and $\widetilde{\mathcal{O}}(d^4)$ additional 1-qubit and 2-qubit elementary quantum gates. On the other side, a comprehensive empirical study suggests that representative state-of-the-art classical optimization algorithms/solvers (including Gurobi) would require a super-polynomial time to solve such optimization instances.

quant-ph↗

arXiv2020

On the Degree of Boolean Functions as Polynomials over $\mathbb{Z}_m$

Polynomial representations of Boolean functions over various rings such as $\mathbb{Z}$ and $\mathbb{Z}_m$ have been studied since Minsky and Papert (1969). From then on, they have been employed in a large variety of fields including communication complexity, circuit complexity, learning theory, coding theory and so on. For any integer $m\ge2$, each Boolean function has a unique multilinear polynomial representation over ring $\mathbb Z_m$. The degree of such polynomial is called modulo-$m$ degree, denoted as $\mathrm{deg}_m(\cdot)$. In this paper, we investigate the lower bound of modulo-$m$ degree of Boolean functions. When $m=p^k$ ($k\ge 1$) for some prime $p$, we give a tight lower bound that $\mathrm{deg}_m(f)\geq k(p-1)$ for any non-degenerated function $f:\{0,1\}^n\to\{0,1\}$, provided that $n$ is sufficient large. When $m$ contains two different prime factors $p$ and $q$, we give a nearly optimal lower bound for any symmetric function $f:\{0,1\}^n\to\{0,1\}$ that $\mathrm{deg}_m(f) \geq \frac{n}{2+\frac{1}{p-1}+\frac{1}{q-1}}$.

cs.CC↗

arXiv2019

Hybrid Decision Trees: Longer Quantum Time is Strictly More Powerful

In this paper, we introduce the hybrid query complexity, denoted as $\mathrm{Q}(f;q)$, which is the minimal query number needed to compute $f$, when a classical decision tree is allowed to call $q'$-query quantum subroutines for any $q'\leq q$. We present the following results: $\bullet$ There exists a total Boolean function $f$ such that $\mathrm{Q}(f;1) = \widetilde{\mathcal{O}}(\mathrm{R}(f)^{4/5})$. $\bullet$ $\mathrm{Q}(f;q) = Ω(\mathrm{bs}(f)/q + \sqrt{\mathrm{bs}(f)})$ for any Boolean function $f$; the lower bound is tight when $f$ is the ${\rm O{\small R}}$ function. $\bullet$ $\mathrm{Q}(g \circ {\rm X{\small OR}}_{C \log n};1) = \widetildeΩ(\sqrt{n})$ for some sufficiently large constant $C$, where $g := {\rm B{\small OOL}S{\small IMON}}_n$ is a variant of Simon's problem. Note that $\mathrm{Q}(g\circ {\rm X{\small OR}}_{C \log n}) = \mathcal{O}(\mathrm{polylog}\; n)$. Therefore an exponential separation is established. Furthermore, this open the road to prove the conjecture $\forall k,\,\mathrm{Q}(g \circ {\rm X{\small OR}}_{C \log^{k+1} n};\log^{k} n) = \widetildeΩ(\sqrt{n})$, which would imply the oracle separation $\mathsf{HP}(\mathsf{QSIZE}(n^α))^\mathfrak{O} \subsetneq \mathsf{BQP}^\mathfrak{O}$ for any $α$, where $\mathsf{HP}(\mathsf{QSIZE}(n^α))$ is a complexity class that contains $\mathsf{BQTIME}(n^α)^{\mathsf{BPP}}$ and $\mathsf{BPP}^{\mathsf{BQTIME}(n^α)}$ in any relativized world.

cs.CC↗

arXiv2018

An Active-Passive Measurement Study of TCP Performance over LTE on High-speed Rails

High-speed rail (HSR) systems potentially provide a more efficient way of door-to-door transportation than airplane. However, they also pose unprecedented challenges in delivering seamless Internet service for on-board passengers. In this paper, we conduct a large-scale active-passive measurement study of TCP performance over LTE on HSR. Our measurement targets the HSR routes in China operating at above 300 km/h. We performed extensive data collection through both controlled setting and passive monitoring, obtaining 1732.9 GB data collected over 135719 km of trips. Leveraging such a unique dataset, we measure important performance metrics such as TCP goodput, latency, loss rate, as well as key characteristics of TCP flows, application breakdown, and users' behaviors. We further quantitatively study the impact of frequent cellular handover on HSR networking performance, and conduct in-depth examination of the performance of two widely deployed transport-layer protocols: TCP CUBIC and TCP BBR. Our findings reveal the performance of today's commercial HSR networks "in the wild", as well as identify several performance inefficiencies, which motivate us to design a simple yet effective congestion control algorithm based on BBR to further boost the throughput by up to 36.5%. They together highlight the need to develop dedicated protocol mechanisms that are friendly to extreme mobility.

cs.NI↗

arXiv2018

The Complexity of $(Δ+ 1)$Coloring inCongested Clique, Massively Parallel Computation,and Centralized Local Computation

We present new randomized algorithms that improve the complexity of the classic $(Δ+1)$-coloring problem, and its generalization $(Δ+1)$-list-coloring, in three well-studied models of distributed, parallel, and centralized computation: Distributed Congested Clique: We present an $O(1)$-round randomized algorithm for $(Δ+1)$-list coloring in the congested clique model of distributed computing. This settles the asymptotic complexity of this problem. It moreover improves upon the $O(\log^\ast Δ)$-round randomized algorithms of Parter and Su [DISC'18] and $O((\log\log Δ)\cdot \log^\ast Δ)$-round randomized algorithm of Parter [ICALP'18]. Massively Parallel Computation: We present a $(Δ+1)$-list coloring algorithm with round complexity $O(\sqrt{\log\log n})$ in the Massively Parallel Computation (MPC) model with strongly sublinear memory per machine. This algorithm uses a memory of $O(n^α)$ per machine, for any desirable constant $α>0$, and a total memory of $\widetilde{O}(m)$, where $m$ is the size of the graph. Notably, this is the first coloring algorithm with sublogarithmic round complexity, in the sublinear memory regime of MPC. For the quasilinear memory regime of MPC, an $O(1)$-round algorithm was given very recently by Assadi et al. [SODA'19]. Centralized Local Computation: We show that $(Δ+1)$-list coloring can be solved with $Δ^{O(1)} \cdot O(\log n)$ query complexity, in the centralized local computation model. The previous state-of-the-art for $(Δ+1)$-list coloring in the centralized local computation model are based on simulation of known LOCAL algorithms.

cs.DS↗

arXiv2017

A Rudimentary Model for Low-Latency Anonymous Communication Systems

In this paper we present a rudimentary model for low-latency anonymous communication systems. Specifically, we study distributed OR algorithm as an abstract of the system. Based on our model, we give several satisfactory lower bounds of anonymity leakage of a deterministic OR algorithm. Some of them reveal a trade-off between anonymity and communication complexity. For the randomized OR algorithm, we only give a relatively trivial but possibly tight lower bound when leaving out communication complexity. And we find the relationship between our model and some open case in the study of secret sharing scheme, if considering communication complexity.

cs.DC↗