arXiv subjects

Fan Wu

Publications and source records attributed to Fan Wu.

At least 199 records · Page 11Linked to original sources

arXiv2025

Enhancement of sensitivity near exceptional points in dissipative qubit-resonator systems

Dissipation usually plays a negative role in quantum metrological technologies, which aim to improve measurement precision by leveraging quantum effects that are vulnerable to environment-induced decoherence. Recently, it has been demonstrated that dissipation can actually be used as a favorable resource for enhancing the susceptibility of signal detection. However, demonstrations of such enhancement for detecting physical quantities in open quantum systems are still lacking. Here we propose and demonstrate a protocol for realizing such non-Hermitian quantum sensors for probing the coupling between a qubit and a resonator subjecting to energy dissipations. The excitation-number conversion associated with the no-jump evolution trajectory enables removal of the noisy outcomes with quantum jumps, implementing the exceptional point (EP), where the Rabi splitting exhibits a divergent behavior in response to a tiny variation of the effective coupling. The sensitivity enhancement near the EP is confirmed by both theoretical calculation and experimental measurement.

quant-ph↗

arXiv2025

Study of the gamma-Ray Radiation Properties of High-redshift Blazars at z>2.5

We study a sample of 30 high-redshift blazars ($z>2.5$) by means of spectra and the radiation mechanism with Fermi Large Area Telescope $γ$-ray observations spanning 15 years. Three models -- the power law, power law with an exponential cutoff, and log-parabola -- are employed to analyze the spectral properties, and most sources exhibit significant curvature. The high-redshift blazars exhibit higher $γ$-ray luminosities and softer spectral indices compared with their low-redshift counterparts, where B3~1343+451 has the highest integrated flux, $\rm 1.13 \times 10^{-7} \mathrm{\ ph \ cm^{-2} s^{-1}}$. We use a standard one-zone leptonic emission model to reproduce the spectral energy distributions of 23 sources with multiwavelength observations. We find that modeling with infrared seed photons is systematically better than with broad-line region (BLR) photons based on a $χ^2$ test, which suggests that the $γ$-ray-emitting regions are most likely located outside the BLR. The fit results show that high-redshift blazars exhibit higher energy density, jet power, kinetic power, and accretion disk luminosities, along with lower synchrotron and inverse Compton (IC) peak frequencies, relative to their lower-redshift counterparts. We find that blazars with higher accretion disk luminosities tend to have lower IC peak frequencies, leading to more efficient cooling of high-energy electrons. The positive correlation between jet power and accretion disk luminosity further supports the possibility of an accretion-jet connection in these high-redshift sources.

astro-ph.HE↗

arXiv2025

MapColorAI: Designing Contextually Relevant Choropleth Map Color Schemes Using a Large Language Model

Choropleth maps, which utilize color schemes to visualize spatial patterns and trends, are simple yet effective tools for geographic data analysis. As such, color scheme design is a critical aspect of choropleth map creation. The traditional coloring methods offered by GIS tools such as ArcGIS and QGIS are not user-friendly for non-professionals. On the one hand, these tools provide numerous color schemes, making it hard to decide which one best matches the theme. On the other hand, it is difficult to fulfill some ambiguous and personalized coloring needs of users, such as requests for 'summer-like' map colors. To address these shortcomings, we develop a novel system that leverages a large language model and map color design principles to generate contextually relevant and user-aligned choropleth map color schemes. The system follows a three-stage process: Data processing, which provides an overview of the data and classifies the data into meaningful classes; Color Concept Design, where the color theme and color mode are conceptualized based on data characteristics and user intentions; and Color Scheme Design, where specific colors are assigned to classes based on generated color theme, color mode, and user requirements. Our system incorporates an interactive interface, providing necessary visualization for choropleth map color design and allowing users to customize and refine color choices flexibly. Through user studies and evaluations, the system demonstrates acceptable usability, accuracy, and flexibility, with users highlighting the tool's efficiency and ease of use.

cs.HC↗

arXiv2025

OpenMLDB: A Real-Time Relational Data Feature Computation System for Online ML

Efficient and consistent feature computation is crucial for a wide range of online ML applications. Typically, feature computation is divided into two distinct phases, i.e., offline stage for model training and online stage for model serving. These phases often rely on execution engines with different interface languages and function implementations, causing significant inconsistencies. Moreover, many online ML features involve complex time-series computations (e.g., functions over varied-length table windows) that differ from standard streaming and analytical queries. Existing data processing systems (e.g., Spark, Flink, DuckDB) often incur multi-second latencies for these computations, making them unsuitable for real-time online ML applications that demand timely feature updates. This paper presents OpenMLDB, a feature computation system deployed in 4Paradigm's SageOne platform and over 100 real scenarios. Technically, OpenMLDB first employs a unified query plan generator for consistent computation results across the offline and online stages, significantly reducing feature deployment overhead. Second, OpenMLDB provides an online execution engine that resolves performance bottlenecks caused by long window computations (via pre-aggregation) and multi-table window unions (via data self-adjusting). It also provides a high-performance offline execution engine with window parallel optimization and time-aware data skew resolving. Third, OpenMLDB features a compact data format and stream-focused indexing to maximize memory usage and accelerate data access. Evaluations in testing and real workloads reveal significant performance improvements and resource savings compared to the baseline systems. The open community of OpenMLDB now has over 150 contributors and gained 1.6k stars on GitHub.

cs.DB↗

arXiv2025

Extrapolated Hard Thresholding Algorithms with Finite Length for Composite $\ell_0$ Penalized Problems

For a class of sparse optimization problems with the penalty function of $\|(\cdot)_+\|_0$, we first characterize its local minimizers and then propose an extrapolated hard thresholding algorithm to solve such problems. We show that the iterates generated by the proposed algorithm with $ε>0$ (where $ε$ is the dry friction coefficient) have finite length, without relying on the Kurdyka-Łojasiewicz inequality. Furthermore, we demonstrate that the algorithm converges to an $ε$-local minimizer of this problem. For the special case that $ε=0$, we establish that any accumulation point of the iterates is a local minimizer of the problem. Additionally, we analyze the convergence when an error term is present in the algorithm, showing that the algorithm still converges in the same manner as before, provided that the errors asymptotically approach zero. Finally, we conduct numerical experiments to verify the theoretical results of the proposed algorithm.

math.OC↗

arXiv2025

A consensus-based optimization method for nonsmooth nonconvex programs with approximated gradient descent scheme

In this paper, we are interested in finding the global minimizer of a nonsmooth nonconvex unconstrained optimization problem. By combining the discrete consensus-based optimization (CBO) algorithm and the gradient descent method, we develop a novel CBO algorithm with an extra gradient descent scheme evaluated by the forward-difference technique on the function values, where only the objective function values are used in the proposed algorithm. First, we prove that the proposed algorithm can exhibit global consensus in an exponential rate in two senses and possess a unique global consensus point. Second, we evaluate the error estimate between the objective function value on the global consensus point and its global minimum. In particular, as the parameter $β$ tends to $\infty$, the error converges to zero and the convergence rate is $\mathcal{O}\left(\frac{\logβ}β\right)$. Third, under some suitable assumptions on the objective function, we provide the number of iterations required for the mean square error in expectation to reach the desired accuracy. It is worth underlining that the theoretical analysis in this paper does not use the mean-field limit. Finally, we illustrate the improved efficiency and promising performance of our novel CBO method through some experiments on several nonconvex benchmark problems and the application to train deep neural networks.

math.OC↗

arXiv2025

Personalized Language Model Learning on Text Data Without User Identifiers

In many practical natural language applications, user data are highly sensitive, requiring anonymous uploads of text data from mobile devices to the cloud without user identifiers. However, the absence of user identifiers restricts the ability of cloud-based language models to provide personalized services, which are essential for catering to diverse user needs. The trivial method of replacing an explicit user identifier with a static user embedding as model input still compromises data anonymization. In this work, we propose to let each mobile device maintain a user-specific distribution to dynamically generate user embeddings, thereby breaking the one-to-one mapping between an embedding and a specific user. We further theoretically demonstrate that to prevent the cloud from tracking users via uploaded embeddings, the local distributions of different users should either be derived from a linearly dependent space to avoid identifiability or be close to each other to prevent accurate attribution. Evaluation on both public and industrial datasets using different language models reveals a remarkable improvement in accuracy from incorporating anonymous user embeddings, while preserving real-time inference requirement.

cs.LG↗

arXiv2025

Dynamic generation of superflow in a fermionic ring through phase imprinting

We study the dynamic generation of persistent current by phase imprinting fermionic atoms in a ring geometry. Mediated by the pairing interaction, the Fermi condensate dynamically acquires a quantized current by developing azimuthal phase slips, as well as density and pairing-order-parameter depletions. Resorting to the Bogolioubov-de Gennes formalism, we investigate the time evolution of the transferred total angular momentum and the quantized superfluid current throughout the phase-imprinting process. This enables a detailed analysis of the impact of interaction, as well as different initial pairing states, on the superflow formation. In particular, we show that, as the condensate is tuned toward the Bose-Einstein-condensate side of the Feshbach resonance, the azimuthal density distribution becomes less susceptible to the phase imprinting potential, leading to smaller quantized current under the same imprinting parameters. Our results offer microscopic insights into the dynamic development of superflow in the phase-imprinting process, and are helpful for the ongoing experimental effort.

cond-mat.quant-gas↗

arXiv2025

Forward Once for All: Structural Parameterized Adaptation for Efficient Cloud-coordinated On-device Recommendation

In cloud-centric recommender system, regular data exchanges between user devices and cloud could potentially elevate bandwidth demands and privacy risks. On-device recommendation emerges as a viable solution by performing reranking locally to alleviate these concerns. Existing methods primarily focus on developing local adaptive parameters, while potentially neglecting the critical role of tailor-made model architecture. Insights from broader research domains suggest that varying data distributions might favor distinct architectures for better fitting. In addition, imposing a uniform model structure across heterogeneous devices may result in risking inefficacy on less capable devices or sub-optimal performance on those with sufficient capabilities. In response to these gaps, our paper introduces Forward-OFA, a novel approach for the dynamic construction of device-specific networks (both structure and parameters). Forward-OFA employs a structure controller to selectively determine whether each block needs to be assembled for a given device. However, during the training of the structure controller, these assembled heterogeneous structures are jointly optimized, where the co-adaption among blocks might encounter gradient conflicts. To mitigate this, Forward-OFA is designed to establish a structure-guided mapping of real-time behaviors to the parameters of assembled networks. Structure-related parameters and parallel components within the mapper prevent each part from receiving heterogeneous gradients from others, thus bypassing the gradient conflicts for coupled optimization. Besides, direct mapping enables Forward-OFA to achieve adaptation through only one forward pass, allowing for swift adaptation to changing interests and eliminating the requirement for on-device backpropagation. Experiments on real-world datasets demonstrate the effectiveness and efficiency of Forward-OFA.

cs.DC↗

arXiv2025

AdaSkip: Adaptive Sublayer Skipping for Accelerating Long-Context LLM Inference

Long-context large language models (LLMs) inference is increasingly critical, motivating a number of studies devoted to alleviating the substantial storage and computational costs in such scenarios. Layer-wise skipping methods are promising optimizations but rarely explored in long-context inference. We observe that existing layer-wise skipping strategies have several limitations when applied in long-context inference, including the inability to adapt to model and context variability, disregard for sublayer significance, and inapplicability for the prefilling phase. This paper proposes \sysname, an adaptive sublayer skipping method specifically designed for long-context inference. \sysname adaptively identifies less important layers by leveraging on-the-fly similarity information, enables sublayer-wise skipping, and accelerates both the prefilling and decoding phases. The effectiveness of \sysname is demonstrated through extensive experiments on various long-context benchmarks and models, showcasing its superior inference performance over existing baselines.

cs.CL↗

arXiv2025

Language Models for Code Optimization: Survey, Challenges and Future Directions

Language models (LMs) built upon deep neural networks (DNNs) have recently demonstrated breakthrough effectiveness in software engineering tasks such as code generation, completion, and repair. This has paved the way for the emergence of LM-based code optimization techniques, which are crucial for enhancing the performance of existing programs, such as accelerating program execution time. However, a comprehensive survey dedicated to this specific application has been lacking. To fill this gap, we present a systematic literature review of over 50 primary studies, identifying emerging trends and addressing 11 specialized questions. Our findings reveal five critical open challenges, such as balancing model complexity with practical usability, cross-language/performance generalizability, and building trust in AI-driven solutions. Furthermore, we provide eight future research directions to facilitate more efficient, robust, and reliable LM-based code optimization. Thereby, this study aims to provide actionable insights and foundational references for both researchers and practitioners in this rapidly evolving field.

cs.SE↗

arXiv2024

Empowering the Edge Intelligence by Air-Ground Integrated Federated Learning

Ubiquitous intelligence has been widely recognized as a critical vision of the future sixth generation (6G) networks, which implies the intelligence over the whole network from the core to the edge including end devices. Nevertheless, fulfilling such vision, particularly the intelligence at the edge, is extremely challenging, due to the limited resources of edge devices as well as the ubiquitous coverage envisioned by 6G. To empower the edge intelligence, in this article, we propose a novel framework called AGIFL (Air-Ground Integrated Federated Learning), which organically integrates air-ground integrated networks and federated learning (FL). In the AGIFL, leveraging the flexible on-demand 3D deployment of aerial nodes such as unmanned aerial vehicles (UAVs), all the nodes can collaboratively train an effective learning model by FL. We also conduct a case study to evaluate the effect of two different deployment schemes of the UAV over the learning and network performance. Last but not the least, we highlight several technical challenges and future research directions in the AGIFL.

cs.NI↗

arXiv2024

MergeNet: Knowledge Migration across Heterogeneous Models, Tasks, and Modalities

In this study, we focus on heterogeneous knowledge transfer across entirely different model architectures, tasks, and modalities. Existing knowledge transfer methods (e.g., backbone sharing, knowledge distillation) often hinge on shared elements within model structures or task-specific features/labels, limiting transfers to complex model types or tasks. To overcome these challenges, we present MergeNet, which learns to bridge the gap of parameter spaces of heterogeneous models, facilitating the direct interaction, extraction, and application of knowledge within these parameter spaces. The core mechanism of MergeNet lies in the parameter adapter, which operates by querying the source model's low-rank parameters and adeptly learning to identify and map parameters into the target model. MergeNet is learned alongside both models, allowing our framework to dynamically transfer and adapt knowledge relevant to the current stage, including the training trajectory knowledge of the source model. Extensive experiments on heterogeneous knowledge transfer demonstrate significant improvements in challenging settings, where representative approaches may falter or prove less applicable.

cs.LG↗

arXiv2024

Dynaseal: A Backend-Controlled LLM API Key Distribution Scheme with Constrained Invocation Parameters

Due to the exceptional performance of Large Language Models (LLMs) in diverse downstream tasks,there has been an exponential growth in edge-device requests to cloud-based models.However, the current authentication mechanism using static Bearer Tokens in request headersfails to provide the flexibility and backend control required for edge-device deployments.To address these limitations, we propose Dynaseal,a novel methodology that enables fine-grained backend constraints on model invocations.

cs.CR↗

arXiv2024

How Does the Smoothness Approximation Method Facilitate Generalization for Federated Adversarial Learning?

Federated Adversarial Learning (FAL) is a robust framework for resisting adversarial attacks on federated learning. Although some FAL studies have developed efficient algorithms, they primarily focus on convergence performance and overlook generalization. Generalization is crucial for evaluating algorithm performance on unseen data. However, generalization analysis is more challenging due to non-smooth adversarial loss functions. A common approach to addressing this issue is to leverage smoothness approximation. In this paper, we develop algorithm stability measures to evaluate the generalization performance of two popular FAL algorithms: \textit{Vanilla FAL (VFAL)} and {\it Slack FAL (SFAL)}, using three different smooth approximation methods: 1) \textit{Surrogate Smoothness Approximation (SSA)}, (2) \textit{Randomized Smoothness Approximation (RSA)}, and (3) \textit{Over-Parameterized Smoothness Approximation (OPSA)}. Based on our in-depth analysis, we answer the question of how to properly set the smoothness approximation method to mitigate generalization error in FAL. Moreover, we identify RSA as the most effective method for reducing generalization error. In highly data-heterogeneous scenarios, we also recommend employing SFAL to mitigate the deterioration of generalization performance caused by heterogeneity. Based on our theoretical results, we provide insights to help develop more efficient FAL algorithms, such as designing new metrics and dynamic aggregation rules to mitigate heterogeneity.

cs.LG↗

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

Contextual Generative Auction with Permutation-level Externalities for Online Advertising

Online advertising has become a core revenue driver for the internet industry, with ad auctions playing a crucial role in ensuring platform revenue and advertiser incentives. Traditional auction mechanisms, like GSP, rely on the independent CTR assumption and fail to account for the influence of other displayed items, termed externalities. Recent advancements in learning-based auctions have enhanced the encoding of high-dimensional contextual features. However, existing methods are constrained by the "allocation-after-prediction" design paradigm, which models set-level externalities within candidate ads and fails to consider the sequential context of the final allocation, leading to suboptimal results. This paper introduces the Contextual Generative Auction (CGA), a novel framework that incorporates permutation-level externalities in multi-slot ad auctions. Built on the structure of our theoretically derived optimal solution, CGA decouples the optimization of allocation and payment. We construct an autoregressive generative model for allocation and reformulate the incentive compatibility (IC) constraint into minimizing ex-post regret that supports gradient computation, enabling end-to-end learning of the optimal payment rule. Extensive offline and online experiments demonstrate that CGA significantly enhances platform revenue and CTR compared to existing methods, while effectively approximating the optimal auction with nearly maximal revenue and minimal regret.

cs.GT↗

arXiv2024

PAR: Prompt-Aware Token Reduction Method for Efficient Large Multimodal Models

Multimodal large language models (MLLMs) demonstrate strong performance across visual tasks, but their efficiency is hindered by significant computational and memory demands from processing long contexts in multimodal inputs. To address this, we introduce PAR (Prompt-Aware Token Reduction), a novel and plug-and-play approach that reduces visual tokens efficiently without compromising model performance. Unlike previous methods that rely heavily on attention mechanisms and overlooking cross-modal interactions , we uses a prompt-aware strategy to adpative identify and cluster essential visual tokens. PAR categorizes visual context redundancy into two types: external and internal. External redundancy is minimized through semantic retrieval, while internal redundancy is addressed using a token routing mechanism. This method substantially reduces computational load without requiring additional training or complex architectural modifications. \textbf{Experimental results demonstrate that across various visual question answering tasks, PAR reduces FLOPs by 83\% with a compression ratio of 89\%, while retaining 97\% of baseline accuracy.} The adaptive design of PAR achieves a 2x token reduction ratio compared to prior approaches, enabling a better balance between performance and efficiency.

cs.CV↗