arXiv subjects

Yunshi Wen

Publications and source records attributed to Yunshi Wen.

5 recordsLinked to original sources

arXiv2026

SC-JEPA: Stabilizing Latent Predictive Learning for Time-Series Anomaly Prediction

Time-series anomaly prediction aims to forecast future system failures before they fully emerge, making latent predictive models such as JEPA a promising framework for capturing precursor dynamics. However, directly applying continuous self-distillation to time-series data is often unstable and can lead to representation collapse, while also struggling to model precursors evolving at different temporal scales. To address this, we propose \textbf{SC-JEPA}, a new JEPA-based framework to model time-series anomaly prediction in a discretized predictive state space. It introduces a soft codebook bottleneck to stabilize latent predictive learning and encourage regime-level structure in the learned representations. Building on this stabilized latent space, we further design a multi-resolution predictive objective to capture precursor patterns at different temporal scales. Experiments on five real-world benchmarks show that SC-JEPA achieves strong and consistent early-warning performance.

cs.LG↗

arXiv2026

CAAD: Causality-Aware Multivariate Time Series Anomaly Detection via Multi-Scale Alignment and Structural Causal Consistency

The operational integrity of complex industrial systems relies on precise anomaly detection and diagnosis. The vast majority of existing methods narrowly focus on capturing temporal similarities of representations, often overlooking the disruption of internal causal relationships, which characterizes system failures and latent anomalies. In this paper, we propose a novel framework (CAAD) that reframes anomaly detection as the continuous verification of Granger causality consistency through exogenous variables. Specifically, the CAAD framework models exogenous time-series variables as residuals, identifying anomalies as significant deviations caused by external interventions. The proposed framework leverages multi-scale alignment to internalize system dynamics and utilizes a gradient-based matrix to monitor internal causal relationship breakdowns. By quantifying causal deviations of both dynamic evolution and relational topology, the CAAD is able to capture subtle causal shifts to achieve precise anomaly detection. Extensive experiments on real-world industrial datasets demonstrate that the CAAD achieves high-precision anomaly detection, outperforming most state-of-the-art baselines.

cs.LG↗

arXiv2026

Revisiting the Generic Transformer: Deconstructing a Strong Baseline for Time Series Foundation Models

The recent surge in Time Series Foundation Models has rapidly advanced the field, yet the heterogeneous training setups across studies make it difficult to attribute improvements to architectural innovations versus data engineering. In this work, we investigate the potential of a standard patch Transformer, demonstrating that this generic architecture achieves state-of-the-art zero-shot forecasting performance using a straightforward training protocol. We conduct a comprehensive ablation study that covers model scaling, data composition, and training techniques to isolate the essential ingredients for high performance. Our findings identify the key drivers of performance, while confirming that the generic architecture itself demonstrates excellent scalability. By strictly controlling these variables, we provide comprehensive empirical results on model scaling across multiple dimensions. We release our open-source model and detailed findings to establish a transparent, reproducible baseline for future research.

cs.LG↗

arXiv2025

Abstracted Shapes as Tokens -- A Generalizable and Interpretable Model for Time-series Classification

In time-series analysis, many recent works seek to provide a unified view and representation for time-series across multiple domains, leading to the development of foundation models for time-series data. Despite diverse modeling techniques, existing models are black boxes and fail to provide insights and explanations about their representations. In this paper, we present VQShape, a pre-trained, generalizable, and interpretable model for time-series representation learning and classification. By introducing a novel representation for time-series data, we forge a connection between the latent space of VQShape and shape-level features. Using vector quantization, we show that time-series from different domains can be described using a unified set of low-dimensional codes, where each code can be represented as an abstracted shape in the time domain. On classification tasks, we show that the representations of VQShape can be utilized to build interpretable classifiers, achieving comparable performance to specialist models. Additionally, in zero-shot learning, VQShape and its codebook can generalize to previously unseen datasets and domains that are not included in the pre-training process. The code and pre-trained weights are available at https://github.com/YunshiWen/VQShape.

cs.LG↗

arXiv2023

High-Speed High-Accuracy Spatial Curve Tracking Using Motion Primitives in Industrial Robots

Industrial robots are increasingly deployed in applications requiring an end effector tool to closely track a specified path, such as in spraying and welding. Performance and productivity present possibly conflicting objectives: tracking accuracy, path speed, and motion uniformity. Industrial robots are programmed through motion primitives consisting of waypoints connected by pre-defined motion segments, with specified parameters such as path speed and blending zone. The actual executed robot motion depends on the robot joint servo controller and joint motion constraints (velocity, acceleration, etc.) which are largely unknown to the users. Programming a robot to achieve the desired performance today is time-consuming and mostly manual, requiring tuning a large number of coupled parameters in the motion primitives. The performance also depends on the choice of additional parameters: possible redundant degrees of freedom, location of the target curve, and the robot configuration. This paper presents a systematic approach to optimize the robot motion primitives for performance. The approach first selects the static parameters, then the motion primitives, and finally iteratively update the waypoints to minimize the tracking error. The ultimate performance objective is to maximize the path speed subject to the tracking accuracy and speed uniformity constraints over the entire path. We have demonstrated the effectiveness of this approach in simulation for ABB and FANUC robots for two challenging example curves, and experimentally for an ABB robot. Comparing with the baseline using the current industry practice, the optimized performance shows over 200% performance improvement.

cs.RO↗