arXiv subjects

Xin Wang

Publications and source records attributed to Xin Wang.

At least 667 records · Page 37Linked to original sources

arXiv2025

Efficient Large-Scale Quantum Optimization via Counterdiabatic Ansatz

Quantum Approximate Optimization Algorithm (QAOA) is one of the fundamental variational quantum algorithms, while a version of QAOA that includes counterdiabatic driving, termed Digitized Counterdiabatic QAOA (DC-QAOA), is generally considered to outperform QAOA for all system sizes when the circuit depth for the two algorithms are held equal. Nevertheless, DC-QAOA introduces more CNOT gates per layer, so the overall circuit complexity is a tradeoff between the number of CNOT gates per layer and the circuit depth and must be carefully assessed. In this paper, we conduct a comprehensive comparison of DC-QAOA and QAOA on MaxCut problem with the total number of CNOT gates held equal, and we focus on one implementation of counterdiabatic terms using nested commutators in DC-QAOA, termed as DC-QAOA(NC). We have found that DC-QAOA(NC) reduces the overall circuit complexity as compared to QAOA only for sufficiently large problems, and for MaxCut problem the number of qubits must exceed 16 for DC-QAOA(NC) to outperform QAOA. Additionally, we benchmark DC-QAOA(NC) against QAOA on the Sherrington-Kirkpatrick model under realistic noise conditions, finding that DC-QAOA(NC) exhibits significantly improved robustness compared to QAOA, maintaining higher fidelity as the problem size scales. Notably, in a direct comparison between one-layer DC-QAOA(NC) and three-layer QAOA where both use the same number of CNOT gates, we identify an exponential performance advantage for DC-QAOA(NC), further signifying its suitability for large-scale quantum optimization tasks. Moreover, based on our finding, we have devised an instance-sequential training method for DC-QAOA(NC) circuits, which, compared to traditional methods, offers performance improvement while using even fewer quantum resources.

quant-ph↗

arXiv2025

Towards Quantum-Native Communication Systems: State-of-the-Art, Trends, and Challenges

The potential synergy between quantum communications and future wireless communication systems is explored. By proposing a quantum-native or quantum-by-design philosophy, the survey examines technologies such as quantumdomain (QD) multi-input multi-output, QD non-orthogonal multiple access, quantum secure direct communication, QD resource allocation, QD routing, and QD artificial intelligence. The recent research advances in these areas are summarized. Given the behavior of photonic and particle-like Terahertz (THz) systems, a comprehensive system-oriented perspective is adopted to assess the feasibility of using quantum communications in future systems. This survey also reviews quantum optimization algorithms and quantum neural networks to explore the potential integration of quantum communication and quantum computing in future systems. Additionally, the current status of quantum sensing, quantum radar, and quantum timing is briefly reviewed in support of future applications. The associated research gaps and future directions are identified, including extending the entanglement coherence time, developing THz quantum communications devices, addressing challenges in channel estimation and tracking, and establishing the theoretical bounds and performance trade-offs of quantum communication, computing, and sensing. This survey offers a unique perspective on the potential for quantum communications to revolutionize future systems and pave the way for even more advanced technologies.

quant-ph↗

arXiv2025

PhyX: Does Your Model Have the "Wits" for Physical Reasoning?

Existing benchmarks fail to capture a crucial aspect of intelligence: physical reasoning, the integrated ability to combine domain knowledge, symbolic reasoning, and understanding of real-world constraints. To address this gap, we introduce PhyX: the first large-scale benchmark designed to assess models capacity for physics-grounded reasoning in visual scenarios. PhyX includes 3K meticulously curated multimodal questions spanning 6 reasoning types across 25 sub-domains and 6 core physics domains: thermodynamics, electromagnetism, mechanics, modern physics, optics, and wave\&acoustics. In our comprehensive evaluation, even state-of-the-art models struggle significantly with physical reasoning. GPT-4o, Claude3.7-Sonnet, and GPT-o4-mini achieve only 32.5%, 42.2%, and 45.8% accuracy respectively-performance gaps exceeding 29% compared to human experts. Our analysis exposes critical limitations in current models: over-reliance on memorized disciplinary knowledge, excessive dependence on mathematical formulations, and surface-level visual pattern matching rather than genuine physical understanding. We provide in-depth analysis through fine-grained statistics, detailed case studies, and multiple evaluation paradigms to thoroughly examine physical reasoning capabilities. To ensure reproducibility, we implement a compatible evaluation protocol based on widely-used toolkits such as VLMEvalKit, enabling one-click evaluation. More details are available on our project page: https://phyx-bench.github.io/.

cs.AI↗

arXiv2025

Understanding the Information Propagation Effects of Communication Topologies in LLM-based Multi-Agent Systems

The communication topology in large language model-based multi-agent systems fundamentally governs inter-agent collaboration patterns, critically shaping both the efficiency and effectiveness of collective decision-making. While recent studies for communication topology automated design tend to construct sparse structures for efficiency, they often overlook why and when sparse and dense topologies help or hinder collaboration. In this paper, we present a causal framework to analyze how agent outputs, whether correct or erroneous, propagate under topologies with varying sparsity. Our empirical studies reveal that moderately sparse topologies, which effectively suppress error propagation while preserving beneficial information diffusion, typically achieve optimal task performance. Guided by this insight, we propose a novel topology design approach, EIB-leanrner, that balances error suppression and beneficial information propagation by fusing connectivity patterns from both dense and sparse graphs. Extensive experiments show the superior effectiveness, communication cost, and robustness of EIB-leanrner.

cs.MA↗

arXiv2025

Flexible Sampling for Long-tailed Skin Lesion Classification

Most of the medical tasks naturally exhibit a long-tailed distribution due to the complex patient-level conditions and the existence of rare diseases. Existing long-tailed learning methods usually treat each class equally to re-balance the long-tailed distribution. However, considering that some challenging classes may present diverse intra-class distributions, re-balancing all classes equally may lead to a significant performance drop. To address this, in this paper, we propose a curriculum learning-based framework called Flexible Sampling for the long-tailed skin lesion classification task. Specifically, we initially sample a subset of training data as anchor points based on the individual class prototypes. Then, these anchor points are used to pre-train an inference model to evaluate the per-class learning difficulty. Finally, we use a curriculum sampling module to dynamically query new samples from the rest training samples with the learning difficulty-aware sampling probability. We evaluated our model against several state-of-the-art methods on the ISIC dataset. The results with two long-tailed settings have demonstrated the superiority of our proposed training strategy, which achieves a new benchmark for long-tailed skin lesion classification.

cs.CV↗

arXiv2025

Effective Fixed-Time Control for Constrained Nonlinear System

In this paper, we tackle the state transformation problem in non-strict full state-constrained systems by introducing an adaptive fixed-time control method, utilizing a one-to-one asymmetric nonlinear mapping auxiliary system. Additionally, we develop a class of multi-threshold event-triggered control strategies that facilitate autonomous controller updates, substantially reducing communication resource consumption. Notably, the self-triggered strategy distinguishes itself from other strategies by obviating the need for continuous real-time monitoring of the controller's state variables. By accurately forecasting the subsequent activation instance, this strategy significantly optimizes the efficiency of the control system. Moreover, our theoretical analysis demonstrates that the semi-global practical fixed-time stability (SPFTS) criterion guarantees both tracking accuracy and closed-loop stability under state constraints, with convergence time independent of initial conditions. Finally, simulation results reveal that the proposed method significantly decreases the frequency of control command updates while maintaining tracking accuracy.

eess.SY↗

arXiv2025

BurstGPT: A Real-world Workload Dataset to Optimize LLM Serving Systems

Serving systems for Large Language Models (LLMs) are often optimized to improve quality of service (QoS) and throughput. However, due to the lack of open-source LLM serving workloads, these systems are frequently evaluated under unrealistic workload assumptions. Consequently, performance may degrade when systems are deployed in real-world scenarios. This work presents BurstGPT, an LLM serving workload with 10.31 million traces from regional Azure OpenAI GPT services over 213 days. BurstGPT captures LLM serving characteristics from user, model and system perspectives: (1) User request concurrency: burstiness variations of requests in Azure OpenAI GPT services, revealing diversified concurrency patterns in different services and model types. (2) User conversation patterns: counts and intervals within conversations for service optimizations. (3) Model response lengths: auto-regressive serving processes of GPT models, showing statistical relations between requests and their responses. (4) System response failures: failures of conversation and API services, showing intensive resource needs and limited availability of LLM services in Azure. The details of the characteristics can serve multiple purposes in LLM serving optimizations, such as system evaluation and trace provisioning. In our demo evaluation with BurstGPT, frequent variations in BurstGPT reveal declines in efficiency, stability, or reliability in realistic LLM serving. We identify that the generalization of KV cache management, scheduling and disaggregation optimizations can be improved under realistic workload evaluations. BurstGPT is publicly available now at https://github.com/HPMLL/BurstGPT and is widely used to develop prototypes of LLM serving frameworks in the industry.

cs.DC↗

arXiv2025

Optimization-Inspired Few-Shot Adaptation for Large Language Models

Large Language Models (LLMs) have demonstrated remarkable performance in real-world applications. However, adapting LLMs to novel tasks via fine-tuning often requires substantial training data and computational resources that are impractical in few-shot scenarios. Existing approaches, such as in-context learning and Parameter-Efficient Fine-Tuning (PEFT), face key limitations: in-context learning introduces additional inference computational overhead with limited performance gains, while PEFT models are prone to overfitting on the few demonstration examples. In this work, we reinterpret the forward pass of LLMs as an optimization process, a sequence of preconditioned gradient descent steps refining internal representations. Based on this connection, we propose Optimization-Inspired Few-Shot Adaptation (OFA), integrating a parameterization that learns preconditioners without introducing additional trainable parameters, and an objective that improves optimization efficiency by learning preconditioners based on a convergence bound, while simultaneously steering the optimization path toward the flat local minimum. Our method overcomes both issues of ICL-based and PEFT-based methods, and demonstrates superior performance over the existing methods on a variety of few-shot adaptation tasks in experiments.

cs.LG↗

arXiv2025

Preserving AUC Fairness in Learning with Noisy Protected Groups

The Area Under the ROC Curve (AUC) is a key metric for classification, especially under class imbalance, with growing research focus on optimizing AUC over accuracy in applications like medical image analysis and deepfake detection. This leads to fairness in AUC optimization becoming crucial as biases can impact protected groups. While various fairness mitigation techniques exist, fairness considerations in AUC optimization remain in their early stages, with most research focusing on improving AUC fairness under the assumption of clean protected groups. However, these studies often overlook the impact of noisy protected groups, leading to fairness violations in practice. To address this, we propose the first robust AUC fairness approach under noisy protected groups with fairness theoretical guarantees using distributionally robust optimization. Extensive experiments on tabular and image datasets show that our method outperforms state-of-the-art approaches in preserving AUC fairness. The code is in https://github.com/Purdue-M2/AUC_Fairness_with_Noisy_Groups.

cs.LG↗

arXiv2025

Predictive Performance of Deep Quantum Data Re-uploading Models

Quantum machine learning models incorporating data re-uploading circuits have garnered significant attention due to their exceptional expressivity and trainability. However, their ability to generate accurate predictions on unseen data, referred to as the predictive performance, remains insufficiently investigated. This study reveals a fundamental limitation in predictive performance when deep encoding layers are employed within the data re-uploading model. Concretely, we theoretically demonstrate that when processing high-dimensional data with limited-qubit data re-uploading models, their predictive performance progressively degenerates to near random-guessing levels as the number of encoding layers increases. In this context, the repeated data uploading cannot mitigate the performance degradation. These findings are validated through experiments on both synthetic linearly separable datasets and real-world datasets. Our results demonstrate that when processing high-dimensional data, the quantum data re-uploading models should be designed with wider circuit architectures rather than deeper and narrower ones.

quant-ph↗

arXiv2025

You Can Wash Hands Better: Accurate Daily Handwashing Assessment with a Smartwatch

Hand hygiene is among the most effective daily practices for preventing infectious diseases such as influenza, malaria, and skin infections. While professional guidelines emphasize proper handwashing to reduce the risk of viral infections, surveys reveal that adherence to these recommendations remains low. To address this gap, we propose UWash, a wearable solution leveraging smartwatches to evaluate handwashing procedures, aiming to raise awareness and cultivate high-quality handwashing habits. We frame the task of handwashing assessment as an action segmentation problem, similar to those in computer vision, and introduce a simple yet efficient two-stream UNet-like network to achieve this goal. Experiments involving 51 subjects demonstrate that UWash achieves 92.27% accuracy in handwashing gesture recognition, an error of <0.5 seconds in onset/offset detection, and an error of <5 points in gesture scoring under user-dependent settings. The system also performs robustly in user-independent and user-independent-location-independent evaluations. Remarkably, UWash maintains high performance in real-world tests, including evaluations with 10 random passersby at a hospital 9 months later and 10 passersby in an in-the-wild test conducted 2 years later. UWash is the first system to score handwashing quality based on gesture sequences, offering actionable guidance for improving daily hand hygiene. The code and dataset are publicly available at https://github.com/aiotgroup/UWash

eess.SP↗

arXiv2025

Quantum Compiler Design for Qubit Mapping and Routing: A Cross-Architectural Survey of Superconducting, Trapped-Ion, and Neutral Atom Systems

Quantum hardware development is progressing rapidly with substantial advancements achieved across leading platforms, including superconducting circuits, trapped-ion systems, and neutral atom arrays. As the pursuit of practical quantum advantage continues, efficient quantum program compilation becomes essential for transforming high-level representations of quantum algorithms into physically executable circuits. A fundamental challenge in this process is qubit mapping and gate scheduling, which play a critical role in adapting compiled circuits to the architectural constraints and physical limitations of specific quantum hardware. In this survey, we systematically review and categorize research on the qubit mapping and routing problems across the three mainstream quantum hardware platforms. We primarily explore the development of hardware-aware compilers for superconducting platforms, classifying existing methods into solver-based, heuristic-based, and machine learning-based approaches, and analyze their optimization targets, including gate count, circuit duration, fidelity, and scalability. Furthermore, we examine the evolution of trapped-ion and neutral atom devices, analyzing the distinct challenges posed by their hardware characteristics and highlighting specialized compilers tailored to these unique physical constraints. Finally, we summarize the key challenges and identify some promising opportunities for future research in quantum compiler design across these hardware platforms.

quant-ph↗

arXiv2025

Power of quantum measurement in simulating unphysical operations

The manipulation of quantum states through linear maps beyond quantum operations has many important applications in various areas of quantum information processing. Current methods simulate unphysical maps by sampling physical operations according to classically determined probability distributions. In this work, we show that using quantum measurement instead leads to lower simulation costs for general Hermitian-preserving maps. Remarkably, we establish the equality between the simulation cost and the well-known diamond norm, thus closing a previously known gap and assigning diamond norm a universal operational meaning for all Hermitian-preserving maps. We demonstrate our method in two applications closely related to error mitigation and quantum machine learning, where it exhibits a favorable scaling. These findings highlight the power of quantum measurement in simulating unphysical operations, in which quantum interference is believed to play a vital role. Our work paves the way for more efficient sampling techniques and has the potential to be extended to more quantum information processing scenarios.

quant-ph↗

arXiv2025

Galaxy Rest-Frame UV Colors at z ~ 2-4 with HST UVCANDELS

We present an analysis of rest-frame UV colors of 17,243 galaxies at $z\sim2-4$ in the HST UVCANDELS fields: GOODS-N, GOODS-S, COSMOS, and EGS. Here, we study the rest-frame UV spectral slope, $β$, measured via model spectra obtained via spectral energy distribution (SED) fitting, $β_{SED}$, and explore its correlation with various galaxy parameters (photometric redshift, UV magnitude, stellar mass, dust attenuation, star formation rate [SFR], and specific SFR) obtained via SED fitting with Dense Basis. We also obtain measurements for $β$ via photometric power-law fitting and compare them to our SED-fit-based results, finding good agreement on average. While we find little evolution in $β$ with redshift from $z=2-4$ for the full population, there are clear correlations between $β$ (and related parameters) when binned by stellar mass. For this sample, lower stellar mass galaxies (log[$M_*$] = 7.5-8.5 $M_\odot$) are typically bluer ($β_{SED}=-2.0\pm 0.2$ / $β_{PL} = -2.1\pm0.4$), fainter ($MUV = -17.8^{+0.7}_{-0.6}$) less dusty ($A{v}=0.4\pm0.1$ mag), exhibit lower rates of star formation (log[SFR]=$0.1\pm0.2 M_\odot/$ yr) and higher specific star formation rates (log[sSFR]=$-8.2\pm0.2 \ \mathrm{yr}^{-1}$) than their high-mass counterparts. Higher-mass galaxies (log[$M_*$] $=10.0-12.0 \ M_\odot$) are on average redder ($β_{SED}=-0.9^{+0.8}_{-0.5}$ / $β_{PL}=-1.0^{+0.8}_{-0.5}$), brighter ($MUV=-19.6^{+1.0}_{-1.2}$), dustier ($Av = 0.9^{+0.5}_{-0.4}$ mag), have higher SFRs (log[SFR]=$1.2^{+0.6}_{-1.1} M_\odot$ yr), and lower sSFRs (log[sSFR]=$-9.1^{+0.5}_{-1.1} {yr}^{-1}$). This study's substantial sample size provides a benchmark for demonstrating that the rest-frame UV spectral slope correlates with stellar mass-dependent galaxy characteristics at $z\sim2-4$, a relationship less discernible with smaller datasets typically available at higher redshifts.

astro-ph.GA↗

arXiv2025

NESTFUL: A Benchmark for Evaluating LLMs on Nested Sequences of API Calls

The resurgence of autonomous agents built using large language models (LLMs) to solve complex real-world tasks has brought increased focus on LLMs' fundamental ability of tool or function calling. At the core of these agents, an LLM must plan, execute, and respond using external tools, APIs, and custom functions. Research on tool calling has gathered momentum, but evaluation benchmarks and datasets representing the complexity of the tasks have lagged behind. In this work, we focus on one such complexity, nested sequencing, with the goal of extending existing benchmarks and evaluation. Specifically, we present NESTFUL, a benchmark to evaluate LLMs on nested sequences of API calls, i.e., sequences where the output of one API call is passed as input to a subsequent call. NESTFUL contains 1800+ nested sequences where all the function calls are executable. Experimental results on a variety of models show that the best-performing model (GPT-4o) achieves a full sequence match accuracy of 28% and a win-rate of 60%, necessitating a large scope for improvement in the nested sequencing aspect of function calling. Our analysis of these results provides possible future research directions for the community, in addition to a benchmark to track progress. We have released the NESTFUL dataset under the Apache 2.0 license at https://github.com/IBM/NESTFUL.

cs.AI↗

arXiv2025

Systematic Evaluation of Machine-Generated Reasoning and PHQ-9 Labeling for Depression Detection Using Large Language Models

Recent research leverages large language models (LLMs) for early mental health detection, such as depression, often optimized with machine-generated data. However, their detection may be subject to unknown weaknesses. Meanwhile, quality control has not been applied to these generated corpora besides limited human verifications. Our goal is to systematically evaluate LLM reasoning and reveal potential weaknesses. To this end, we first provide a systematic evaluation of the reasoning over machine-generated detection and interpretation. Then we use the models' reasoning abilities to explore mitigation strategies for enhanced performance. Specifically, we do the following: A. Design an LLM instruction strategy that allows for systematic analysis of the detection by breaking down the task into several subtasks. B. Design contrastive few-shot and chain-of-thought prompts by selecting typical positive and negative examples of detection reasoning. C. Perform human annotation for the subtasks identified in the first step and evaluate the performance. D. Identify human-preferred detection with desired logical reasoning from the few-shot generation and use them to explore different optimization strategies. We conducted extensive comparisons on the DepTweet dataset across the following subtasks: 1. identifying whether the speaker is describing their own depression; 2. accurately detecting the presence of PHQ-9 symptoms, and 3. finally, detecting depression. Human verification of statistical outliers shows that LLMs demonstrate greater accuracy in analyzing and detecting explicit language of depression as opposed to implicit expressions of depression. Two optimization methods are used for performance enhancement and reduction of the statistic bias: supervised fine-tuning (SFT) and direct preference optimization (DPO). Notably, the DPO approach achieves significant performance improvement.

cs.CL↗

arXiv2025

SafeVid: Toward Safety Aligned Video Large Multimodal Models

As Video Large Multimodal Models (VLMMs) rapidly advance, their inherent complexity introduces significant safety challenges, particularly the issue of mismatched generalization where static safety alignments fail to transfer to dynamic video contexts. We introduce SafeVid, a framework designed to instill video-specific safety principles in VLMMs. SafeVid uniquely transfers robust textual safety alignment capabilities to the video domain by employing detailed textual video descriptions as an interpretive bridge, facilitating LLM-based rule-driven safety reasoning. This is achieved through a closed-loop system comprising: 1) generation of SafeVid-350K, a novel 350,000-pair video-specific safety preference dataset; 2) targeted alignment of VLMMs using Direct Preference Optimization (DPO); and 3) comprehensive evaluation via our new SafeVidBench benchmark. Alignment with SafeVid-350K significantly enhances VLMM safety, with models like LLaVA-NeXT-Video demonstrating substantial improvements (e.g., up to 42.39%) on SafeVidBench. SafeVid provides critical resources and a structured approach, demonstrating that leveraging textual descriptions as a conduit for safety reasoning markedly improves the safety alignment of VLMMs. We have made SafeVid-350K dataset (https://huggingface.co/datasets/yxwang/SafeVid-350K) publicly available.

cs.CV↗

arXiv2025

Model-Targeted Data Poisoning Attacks against ITS Applications with Provable Convergence

The growing reliance of intelligent systems on data makes the systems vulnerable to data poisoning attacks. Such attacks could compromise machine learning or deep learning models by disrupting the input data. Previous studies on data poisoning attacks are subject to specific assumptions, and limited attention is given to learning models with general (equality and inequality) constraints or lacking differentiability. Such learning models are common in practice, especially in Intelligent Transportation Systems (ITS) that involve physical or domain knowledge as specific model constraints. Motivated by ITS applications, this paper formulates a model-target data poisoning attack as a bi-level optimization problem with a constrained lower-level problem, aiming to induce the model solution toward a target solution specified by the adversary by modifying the training data incrementally. As the gradient-based methods fail to solve this optimization problem, we propose to study the Lipschitz continuity property of the model solution, enabling us to calculate the semi-derivative, a one-sided directional derivative, of the solution over data. We leverage semi-derivative descent to solve the bi-level optimization problem, and establish the convergence conditions of the method to any attainable target model. The model and solution method are illustrated with a simulation of a poisoning attack on the lane change detection using SVM.

math.OC↗