arXiv subjects

Sheng Pan

Publications and source records attributed to Sheng Pan.

7 recordsLinked to original sources

arXiv2026

LeapTS: Rethinking Time Series Forecasting as Adaptive Multi-Horizon Scheduling

Time series forecasting serves as an essential tool for many real-world applications, supporting tasks such as resource optimization and decision-making. Despite significant architectural advancements, most modern models still treat forecasting task as a fixed mapping from history to target horizons. This induces temporal decoupling across future time points and limits the model's ability to adapt to the evolving context as forecasting progresses. In this work, we present LeapTS, a novel framework that reformulates time series forecasting as a dynamic scheduling process over the prediction horizon. Specifically, LeapTS organizes the forecasting process into multi-level decisions using: (1) the hierarchical controller to dynamically select the optimal prediction scale and advancement length at each step, and (2) continuous-time state evolution driven by neural controlled differential equations. Within this process, the controlled update mechanism explicitly couples the irregular temporal dynamics with discrete scheduling feedback. Extensive evaluations on both real-world and synthetic datasets demonstrate that LeapTS improves overall forecasting performance by at least 7.4% while achieving a 2.6$\times$ to 5.3$\times$ inference speedup over representative Transformer-based models. Furthermore, by explicitly tracing the scheduling trajectories, we reveal how the model autonomously adapts its forecasting behavior to capture non-stationary dynamics.

cs.LG↗

arXiv2026

Beyond Passive Aggregation: Active Auditing and Topology-Aware Defense in Decentralized Federated Learning

Decentralized Federated Learning (DFL) remains highly vulnerable to adaptive backdoor attacks designed to bypass traditional passive defense metrics. To address this limitation, we shift the defensive paradigm toward a novel active, interventional auditing framework. First, we establish a dynamical model to characterize the spatiotemporal diffusion of adversarial updates across complex graph topologies. Second, we introduce a suite of proactive auditing metrics, stochastic entropy anomaly, randomized smoothing Kullback-Leibler divergence, and activation kurtosis. These metrics utilize private probes to stress-test local models, effectively exposing latent backdoors that remain invisible to conventional static detection. Furthermore, we implement a topology-aware defense placement strategy to maximize global aggregation resilience. We provide theoretical property for the system's convergence under co-evolving attack and defense dynamics. Numeric empirical evaluations across diverse architectures demonstrate that our active framework is highly competitive with state-of-the-art defenses in mitigating stealthy, adaptive backdoors while preserving primary task utility.

cs.LG↗

arXiv2026

TimeOmni-VL: Unified Models for Time Series Understanding and Generation

Recent time series modeling faces a sharp divide between numerical generation and semantic understanding, with research showing that generation models often rely on superficial pattern matching, while understanding-oriented models struggle with high-fidelity numerical output. Although unified multimodal models (UMMs) have bridged this gap in vision, their potential for time series remains untapped. We propose TimeOmni-VL, the first vision-centric framework that unifies time series understanding and generation through two key innovations: (1) Fidelity-preserving bidirectional mapping between time series and images (Bi-TSI), which advances Time Series-to-Image (TS2I) and Image-to-Time Series (I2TS) conversions to ensure near-lossless transformations. (2) Understanding-guided generation. We introduce TSUMM-Suite, a novel dataset consisting of six understanding tasks rooted in time series analytics and coupled with two generation tasks. With a calibrated Chain-of-Thought, TimeOmni-VL is the first to leverage time series understanding as an explicit control signal for high-fidelity generation. Experiments confirm that this unified approach significantly improves semantic understanding and numerical precision, establishing a new frontier for multimodal time series modeling.

cs.LG↗

arXiv2026

It's TIME: Towards the Next Generation of Time Series Forecasting Benchmarks

Time series foundation models (TSFMs) are revolutionizing the forecasting landscape from specific dataset modeling to generalizable task evaluation. However, we contend that existing benchmarks exhibit common limitations in four dimensions: constrained data composition dominated by reused legacy sources, compromised data integrity lacking rigorous quality assurance, misaligned task formulations detached from real-world contexts, and rigid analysis perspectives that obscure generalizable insights. To bridge these gaps, we introduce TIME, a next-generation task-centric benchmark comprising 50 fresh datasets and 98 forecasting tasks, tailored for strict zero-shot TSFM evaluation free from data leakage. Integrating large language models and human expertise, we establish a human-in-the-loop benchmark construction pipeline to ensure high data integrity and redefine task formulation by aligning forecasting configurations with real-world operational requirements and variate predictability. Furthermore, we propose a novel pattern-level evaluation perspective that moves beyond traditional dataset-level evaluations based on static meta labels. By leveraging structural time series features to characterize intrinsic temporal properties, this approach offers generalizable insights into model capabilities across diverse patterns. We evaluate 12 TSFMs and establish a multi-granular leaderboard to facilitate in-depth analysis and visualized inspection. The leaderboard is available at https://huggingface.co/spaces/Real-TSF/TIME-leaderboard.

cs.LG↗

arXiv2025

Robust transfer regression with corrupted labels

In this paper, we introduce a robust transfer regression method designed to handle corrupted labels in target data, under the scenarios that the corruption affects a substantial portion of the labels and the locations of these corruptions are unknown. Theoretical analysis substantiates our approach, illustrating that the estimation error consists of three components: the first relates to the source data; the second encompasses the domain shift ; and the third captures the estimation error attributed to the corrupted vector. Our theoretical framework ensures that the proposed method surpasses estimations based solely on target data. We validate our method through numerical experiments aimed at reconstructing corrupted compressed signals. Additionally, we apply our method to analyze the association between O6-methylguanine-DNA methyltransferase (MGMT) methylation and gene expression in Glioblastoma (GBM) patients. Keywords:

stat.ME↗

arXiv2024

Finite Volume Physical Informed Neural Network (FV-PINN) with Reduced Derivative Order for Incompressible Flows

Physics-Informed Neural Networks (PINN) has evolved into a powerful tool for solving partial differential equations, which has been applied to various fields such as energy, environment, en-gineering, etc. When utilizing PINN to solve partial differential equations, it is common to rely on Automatic Differentiation (AD) to compute the residuals of the governing equations. This can lead to certain precision losses, thus affecting the accuracy of the network prediction. This paper pro-poses a Finite Volume Physics-Informed Neural Network (FV-PINN), designed to address steady-state problems of incompressible flow. This method divides the solution domain into mul-tiple grids. Instead of calculating the residuals of the Navier-Stokes equations at collocation points within the grid, as is common in traditional PINNs, this approach evaluates them at Gaussian in-tegral points on the grid boundaries using Gauss's theorem. The loss function is constructed using the Gaussian integral method, and the differentiation order for velocity is reduced. To validate the effectiveness of this approach, we predict the velocity and pressure fields for two typical examples in fluid topology optimization. The results are compared with commercial software COMSOL, which indicates that FVI-PINN significantly improves the prediction accuracy of both the velocity and pressure fields while accelerating the training speed of the network.

physics.flu-dyn↗

arXiv2014

Spectrum and Energy Efficiency Evaluation of Two-Tier Femtocell networks With Partially Open Channels

Two-tier femtocell networks is an efficient communication architecture that significantly improves throughput in indoor environments with low power consumption. Traditionally, a femtocell network is usually configured to be either completely open or completely closed in that its channels are either made available to all users or used by its own users only. This may limit network flexibility and performance. It is desirable for owners of femtocell base stations if a femtocell can partially open its channels for external users access. In such scenarios, spectrum and energy efficiency becomes a critical issue in the design of femtocell network protocols and structure. In this paper, we conduct performance analysis for two-tier femtocell networks with partially open channels. In particular, we build a Markov chain to model the channel access in the femtocell network and then derive the performance metrics in terms of the blocking probabilities. Based on stationary state probabilities derived by Markov chain models, spectrum and energy efficiency are modeled and analyzed under different scenarios characterized by critical parameters, including number of femtocells in a macrocell, average number of users, and number of open channels in a femtocell. Numerical and Monte-Carlo (MC) simulation results indicate that the number of open channels in a femtocell has an adverse impact on the spectrum and energy efficiency of two-tier femtocell networks. Results in this paper provide guidelines for trading off spectrum and energy efficiency of two-tier femtocell networks by configuring different numbers of open channels in a femtocell.

cs.NI↗