arXiv subjects

Bin Yang

Publications and source records attributed to Bin Yang.

At least 109 records · Page 6Linked to original sources

arXiv2026

Empowering Time Series Analysis with Large-Scale Multimodal Pretraining

While existing time series foundation models primarily rely on large-scale unimodal pretraining, they lack complementary modalities to enhance time series understanding. Building multimodal foundation models is a natural next step, but it faces key challenges: 1) lack of a unified multimodal pretraining paradigm and large-scale multimodal corpora for time series analysis; 2) how to effectively integrate heterogeneous modalities and enhance model generalization. To address these challenges, we take an early step toward multimodal foundation models for time series analysis. We first propose a multimodal pretraining paradigm that leverages time series with endogenous modalities (derived images and text) and exogenous knowledge (real-world news), providing a comprehensive multi-view perspective for time series analysis. To support this, we develop an automated data construction pipeline to curate MM-TS, the first large-scale multimodal time series dataset spanning six domains, with up to one billion points. Then we propose HORAI, a frequency-enhanced multimodal foundation model. It integrates two core components: the Frequency-enhanced Cross-Modality Encoder and the Time-Frequency Decoder, designed to effectively fuse multimodal features and enhance model generalization across modalities and domains. After pretraining on MM-TS, HORAI achieves state-of-the-art zero-shot performance on time series forecasting and anomaly detection tasks, demonstrating strong generalization.

cs.LG↗

arXiv2026

Learning-based Observer for Coupled Disturbance

Achieving high-precision control for robotic systems is hindered by the low-fidelity dynamical model and external disturbances. Especially, the intricate coupling between internal uncertainties and external disturbances further exacerbates this challenge. This study introduces an effective and convergent algorithm enabling accurate estimation of the coupled disturbance via combining control and learning philosophies. Concretely, by resorting to Chebyshev series expansion, the coupled disturbance is firstly decomposed into an unknown parameter matrix and two known structures dependent on system state and external disturbance respectively. A regularized least squares algorithm is subsequently formalized to learn the parameter matrix using historical time-series data. Finally, a polynomial disturbance observer is specifically devised to achieve a high-precision estimation of the coupled disturbance by utilizing the learned portion. The proposed algorithm is evaluated through extensive simulations and real flight tests. We believe this work can offer a new pathway to integrate learning approaches into control frameworks for addressing longstanding challenges in robotic applications.

cs.RO↗

arXiv2026

D-CORE: Incentivizing Task Decomposition in Large Reasoning Models for Complex Tool Use

Effective tool use and reasoning are essential capabilities for large reasoning models~(LRMs) to address complex real-world problems. Through empirical analysis, we identify that current LRMs lack the capability of sub-task decomposition in complex tool use scenarios, leading to Lazy Reasoning. To address this, we propose a two-stage training framework D-CORE~(\underline{\textbf{D}}ecomposing tasks and \underline{\textbf{Co}}mposing \underline{\textbf{Re}}asoning processes) that first incentivize the LRMs' task decomposition reasoning capability via self-distillation, followed by diversity-aware reinforcement learning~(RL) to restore LRMs' reflective reasoning capability. D-CORE achieves robust tool-use improvements across diverse benchmarks and model scales. Experiments on BFCLv3 demonstrate superiority of our method: D-CORE-8B reaches 77.7\% accuracy, surpassing the best-performing 8B model by 5.7\%. Meanwhile, D-CORE-14B establishes a new state-of-the-art at 79.3\%, outperforming 70B models despite being 5$\times$ smaller. The source code is available at https://github.com/alibaba/EfficientAI.

cs.CL↗

arXiv2026

SEER: Transformer-based Robust Time Series Forecasting via Automated Patch Enhancement and Replacement

Time series forecasting is important in many fields that require accurate predictions for decision-making. Patching techniques, commonly used and effective in time series modeling, help capture temporal dependencies by dividing the data into patches. However, existing patch-based methods fail to dynamically select patches and typically use all patches during the prediction process. In real-world time series, there are often low-quality issues during data collection, such as missing values, distribution shifts, anomalies and white noise, which may cause some patches to contain low-quality information, negatively impacting the prediction results. To address this issue, this study proposes a robust time series forecasting framework called SEER. Firstly, we propose an Augmented Embedding Module, which improves patch-wise representations using a Mixture-of-Experts (MoE) architecture and obtains series-wise token representations through a channel-adaptive perception mechanism. Secondly, we introduce a Learnable Patch Replacement Module, which enhances forecasting robustness and model accuracy through a two-stage process: 1) a dynamic filtering mechanism eliminates negative patch-wise tokens; 2) a replaced attention module substitutes the identified low-quality patches with global series-wise token, further refining their representations through a causal attention mechanism. Comprehensive experimental results demonstrate the SOTA performance of SEER.

cs.LG↗

arXiv2026

A Comprehensive Survey of Deep Learning for Multivariate Time Series Forecasting: A Channel Strategy Perspective

Multivariate Time Series Forecasting (MTSF) plays a crucial role across diverse fields, ranging from economic, energy, to traffic. In recent years, deep learning has demonstrated outstanding performance in MTSF tasks. In MTSF, modeling the correlations among different channels is critical, as leveraging information from other related channels can significantly improve the prediction accuracy of a specific channel. This study systematically reviews the channel modeling strategies for time series and proposes a taxonomy organized into three hierarchical levels: the strategy perspective, the mechanism perspective, and the characteristic perspective. On this basis, we provide a structured analysis of these methods and conduct an in-depth examination of the advantages and limitations of different channel strategies. Finally, we summarize and discuss some future research directions to provide useful research guidance. Moreover, we maintain an up-to-date Github repository (https://github.com/decisionintelligence/CS4TS) which includes all the papers discussed in the survey.

cs.LG↗

arXiv2026

Indoor Positioning Based on Active Radar Sensing and Passive Reflectors: Reflector Placement Optimization

We extend our work on a novel indoor positioning system (IPS) for autonomous mobile robots (AMRs) based on radar sensing of local, passive radar reflectors. Through the combination of simple reflectors and a single-channel frequency modulated continuous wave (FMCW) radar, high positioning accuracy at low system cost can be achieved. Further, a multi-objective (MO) particle swarm optimization (PSO) algorithm is presented that optimizes the 2D placement of radar reflectors in complex room settings.

cs.RO↗

arXiv2026

Investigation of ArUco Marker Placement for Planar Indoor Localization

Indoor localization of autonomous mobile robots (AMRs) can be realized with fiducial markers. Such systems require only a simple, monocular camera as sensor and fiducial markers as passive, identifiable position references that can be printed on a piece of paper and distributed in the area of interest. Thus, fiducial marker systems can be scaled to large areas with a minor increase in system complexity and cost. We investigate the localization behavior of the fiducial marker framework ArUco w.r.t. the placement of the markers including the number of markers, their orientation w.r.t. the camera, and the camera-marker distance. In addition, we propose a simple Kalman filter with adaptive measurement noise variances for real-time AMR tracking.

eess.IV↗

arXiv2026

Post-Perihelion Integral Field Spectroscopy of the Interstellar Comet 3I/ATLAS

The environs of other stellar systems may be directly probed by analyzing the cometary activity of interstellar objects. The recently discovered interstellar object 3I/ATLAS was the subject of an intensive worldwide follow-up campaign in its pre-perihelion approach. Now, 3I/ATLAS has begun its post-perihelion departure from the Solar System. In this letter, we report the first post-perihelion blue-sensitive integral-field unit spectroscopy of 3I/ATLAS using the Keck Cosmic Web Imager on November 16, 2025. We confirm previously reported CN, Fe, and Ni outgassing along with detections of carbon chain molecules $\mathrm{C}_2$ and $\mathrm{C}_3$. We calculate production rates for each species. We find Fe and Ni production rates of $\mathrm{Q_{Fe}} = (9.55\pm3.96)\times10^{25}$ atoms s$^{-1}$, and $\mathrm{Q_{Ni}} = (6.61\pm2.74)\times10^{25}$ atoms s$^{-1}$, resulting in a ratio of $\log(\mathrm{Q_{Ni}} / \mathrm{Q_{Fe}}) = -0.16\pm0.03$, which matches Solar System comets well and continues the pre-perihelion trend of declining $\log(\mathrm{Q_{Ni}} / \mathrm{Q_{Fe}})$ with $r_h$. We investigate the radial distributions of these elemental species and find characteristic $e$-folding radii of 3880$\pm$39 km for Ni, 6053$\pm$68 km for CN, 4194$\pm$45 km for $\mathrm{C}_2$, and 3833$\pm$45 km for $\mathrm{C}_3$. Compared to pre-perihelion measurements, these radii have increased by a factor of $\sim$6.5--7. Our post-perihelion observations reveal that 3I/ATLAS continues to exhibit cometary behavior broadly consistent with Solar System comets.

astro-ph.EP↗

arXiv2026

Test-Time Adaptation For Speech Enhancement Via Mask Polarization

Adapting speech enhancement (SE) models to unseen environments is crucial for practical deployments, yet test-time adaptation (TTA) for SE remains largely under-explored due to a lack of understanding of how SE models degrade under domain shifts. We observe that mask-based SE models lose confidence under domain shifts, with predicted masks becoming flattened and losing decisive speech preservation and noise suppression. Based on this insight, we propose mask polarization (MPol), a lightweight TTA method that restores mask bimodality through distribution comparison using the Wasserstein distance. MPol requires no additional parameters beyond the trained model, making it suitable for resource-constrained edge deployments. Experimental results across diverse domain shifts and architectures demonstrate that MPol achieves very consistent gains that are competitive with significantly more complex approaches.

eess.AS↗

arXiv2026

Task-Aware Mixture-of-Experts for Time Series Analysis

Time Series Analysis is widely used in various real-world applications such as weather forecasting, financial fraud detection, imputation for missing data in IoT systems, and classification for action recognization. Mixture-of-Experts (MoE), as a powerful architecture, though demonstrating effectiveness in NLP, still falls short in adapting to versatile tasks in time series analytics due to its task-agnostic router and the lack of capability in modeling channel correlations. In this study, we propose a novel, general MoE-based time series framework called PatchMoE to support the intricate ``knowledge'' utilization for distinct tasks, thus task-aware. Based on the observation that hierarchical representations often vary across tasks, e.g., forecasting vs. classification, we propose a Recurrent Noisy Gating to utilize the hierarchical information in routing, thus obtaining task-sepcific capability. And the routing strategy is operated on time series tokens in both temporal and channel dimensions, and encouraged by a meticulously designed Temporal \& Channel Load Balancing Loss to model the intricate temporal and channel correlations. Comprehensive experiments on five downstream tasks demonstrate the state-of-the-art performance of PatchMoE.

cs.LG↗

arXiv2026

TimeART: Towards Agentic Time Series Reasoning via Tool-Augmentation

Time series data widely exist in real-world cyber-physical systems. Though analyzing and interpreting them contributes to significant values, e.g, disaster prediction and financial risk control, current workflows mainly rely on human data scientists, which requires significant labor costs and lacks automation. To tackle this, we introduce TimeART, a framework fusing the analytical capability of strong out-of-the-box tools and the reasoning capability of Large Language Models (LLMs), which serves as a fully agentic data scientist for Time Series Question Answering (TSQA). To teach the LLM-based Time Series Reasoning Models (TSRMs) strategic tool-use, we also collect a 100k expert trajectory corpus called TimeToolBench. To enhance TSRMs' generalization capability, we then devise a four-stage training strategy, which boosts TSRMs through learning from their own early experiences and self-reflections. Experimentally, we train an 8B TSRM on TimeToolBench and equip it with the TimeART framework, and it achieves consistent state-of-the-art performance on multiple TSQA tasks, which pioneers a novel approach towards agentic time series reasoning.

cs.LG↗

arXiv2026

ToolCaching: Towards Efficient Caching for LLM Tool-calling

Recent advances in Large Language Models (LLMs) have revolutionized web applications, enabling intelligent search, recommendation, and assistant services with natural language interfaces. Tool-calling extends LLMs with the ability to interact with external APIs, greatly enhancing their practical utility. While prior research has improved tool-calling performance by adopting traditional computer systems techniques, such as parallel and asynchronous execution, the challenge of redundant or repeated tool-calling requests remains largely unaddressed. Caching is a classic solution to this problem, but applying it to LLM tool-calling introduces new difficulties due to heterogeneous request semantics, dynamic workloads, and varying freshness requirements, which render conventional cache policies ineffective. To address these issues, we propose ToolCaching, an efficient feature-driven and adaptive caching framework for LLM tool-calling systems. ToolCaching systematically integrates semantic and system-level features to evaluate request cacheability and estimate caching value. At its core, the VAAC algorithm integrates bandit-based admission with value-driven, multi-factor eviction, jointly accounting for request frequency, recency, and caching value. Extensive experiments on synthetic and public tool-calling workloads demonstrate that ToolCaching with VAAC achieves up to 11% higher cache hit ratios and 34% lower latency compared to standard policies, effectively accelerating LLM tool-calling in practical applications.

cs.SE↗

arXiv2026

Learning to Factorize and Adapt: A Versatile Approach Toward Universal Spatio-Temporal Foundation Models

Spatio-Temporal (ST) Foundation Models (STFMs) promise cross-dataset generalization, yet joint ST pretraining is computationally expensive and grapples with the heterogeneity of domain-specific spatial patterns. Substantially extending our preliminary conference version, we present FactoST-v2, an enhanced factorized framework redesigned for full weight transfer and arbitrary-length generalization. FactoST-v2 decouples universal temporal learning from domain-specific spatial adaptation. The first stage pretrains a minimalist encoder-only backbone using randomized sequence masking to capture invariant temporal dynamics, enabling probabilistic quantile prediction across variable horizons. The second stage employs a streamlined adapter to rapidly inject spatial awareness via meta adaptive learning and prompting. Comprehensive evaluations across diverse domains demonstrate that FactoST-v2 achieves state-of-the-art accuracy with linear efficiency - significantly outperforming existing foundation models in zero-shot and few-shot scenarios while rivaling domain-specific expert baselines. This factorized paradigm offers a practical, scalable path toward truly universal STFMs. Code is available at https://github.com/CityMind-Lab/FactoST.

cs.LG↗

arXiv2026

GMM-COMET: Continual Source-Free Universal Domain Adaptation via a Mean Teacher and Gaussian Mixture Model-Based Pseudo-Labeling

Unsupervised domain adaptation tackles the problem that domain shifts between training and test data impair the performance of neural networks in many real-world applications. Thereby, in realistic scenarios, the source data may no longer be available during adaptation, and the label space of the target domain may differ from the source label space. This setting, known as source-free universal domain adaptation (SF-UniDA), has recently gained attention, but all existing approaches only assume a single domain shift from source to target. In this work, we present the first study on continual SF-UniDA, where the model must adapt sequentially to a stream of multiple different unlabeled target domains. Building upon our previous methods for online SF-UniDA, we combine their key ideas by integrating Gaussian mixture model-based pseudo-labeling within a mean teacher framework for improved stability over long adaptation sequences. Additionally, we introduce consistency losses for further robustness. The resulting method GMM-COMET provides a strong first baseline for continual SF-UniDA and is the only approach in our experiments to consistently improve upon the source-only model across all evaluated scenarios. Our code is available at https://github.com/pascalschlachter/GMM-COMET.

cs.LG↗

arXiv2025

Enhancing Time Series Forecasting through Selective Representation Spaces: A Patch Perspective

Time Series Forecasting has made significant progress with the help of Patching technique, which partitions time series into multiple patches to effectively retain contextual semantic information into a representation space beneficial for modeling long-term dependencies. However, conventional patching partitions a time series into adjacent patches, which causes a fixed representation space, thus resulting in insufficiently expressful representations. In this paper, we pioneer the exploration of constructing a selective representation space to flexibly include the most informative patches for forecasting. Specifically, we propose the Selective Representation Space (SRS) module, which utilizes the learnable Selective Patching and Dynamic Reassembly techniques to adaptively select and shuffle the patches from the contextual time series, aiming at fully exploiting the information of contextual time series to enhance the forecasting performance of patch-based models. To demonstrate the effectiveness of SRS module, we propose a simple yet effective SRSNet consisting of SRS and an MLP head, which achieves state-of-the-art performance on real-world datasets from multiple domains. Furthermore, as a novel plug-and-play module, SRS can also enhance the performance of existing patch-based models. The resources are available at https://github.com/decisionintelligence/SRSNet.

cs.LG↗

arXiv2025

Very Large Telescope Observations of Interstellar Comet 3I/ATLAS. II. From Quiescence to Glow: Dramatic Rise of Ni i Emission and Incipient CN Outgassing at Large Heliocentric Distances*

We report VLT spectroscopy of the interstellar comet 3I/ATLAS (C/2025~N1) from $r_{\rm h}\!\simeq\!4.4$ to $2.85$~au using X-shooter (300--550\,nm, $R\!\simeq\!3000$) and UVES (optical, $R\!\simeq\!35k-80k$). The coma is dust-dominated with a fairly constant red optical continuum slope ($\sim$21--22\%/1000Å). We report detection of CN emission and also detect numerous Ni\,\textsc{ii}~lines while Fe\,\textsc{i}~remains undetected, potentially implying efficiently released gas-phase Ni. At $r_{\rm h}\!\simeq\!3.14$~au we derive $3σ$ limits of $Q({\rm OH})<{1.48\times10^{26}}\ {\rm s^{-1}}$, but find no indications for [O\,\textsc{i}], C$_2$, C$_3$ or NH$_2$. From our latest X-shooter measurements conducted on 2025-08-21 ($r_{\rm h} = 2.85$\,au) we measure production rates of $\log~Q(\mathrm{CN}) = {24.81\pm 0.01}$ molecules s$^{-1}$ and $\log~Q$(Ni) $= {23.30\pm0.07}$ atoms s$^{-1}$, and characterize their evolution as the comet approaches perihelion.~We observe a steep heliocentric-distance scaling for the production rates $Q(\mathrm{Ni}) \propto r_h^{-7.7 \pm 1.0}$ and for $Q(\mathrm{CN}) \propto r_h^{-6.7 \pm 0.2}$, and predict a Ni--CO$_{(2)}$ correlation if the Ni\,\textsc{ii}\ emission is driven by the carbonyl formation channel.~Energetic considerations of activation barriers show that this behavior is inconsistent with direct sublimation of canonical metal/sulfide phases and instead favors low--activation--energy release from dust, e.g.~photon-stimulated desorption or mild thermolysis of metalated organics or Ni-rich nanophases, possibly including Ni--carbonyl-like complexes.~These hypotheses are testable with future coordinated ground-based and space-based monitoring as 3I becomes more active during its continued passage through the solar system.

astro-ph.SR↗

arXiv2025

DOS: Distilling Observable Softmaps of Zipfian Prototypes for Self-Supervised Point Representation

Recent advances in self-supervised learning (SSL) have shown tremendous potential for learning 3D point cloud representations without human annotations. However, SSL for 3D point clouds still faces critical challenges due to irregular geometry, shortcut-prone reconstruction, and unbalanced semantics distribution. In this work, we propose DOS (Distilling Observable Softmaps), a novel SSL framework that self-distills semantic relevance softmaps only at observable (unmasked) points. This strategy prevents information leakage from masked regions and provides richer supervision than discrete token-to-prototype assignments. To address the challenge of unbalanced semantics in an unsupervised setting, we introduce Zipfian prototypes and incorporate them using a modified Sinkhorn-Knopp algorithm, Zipf-Sinkhorn, which enforces a power-law prior over prototype usage and modulates the sharpness of the target softmap during training. DOS outperforms current state-of-the-art methods on semantic segmentation and 3D object detection across multiple benchmarks, including nuScenes, Waymo, SemanticKITTI, ScanNet, and ScanNet200, without relying on extra data or annotations. Our results demonstrate that observable-point softmaps distillation offers a scalable and effective paradigm for learning robust 3D representations.

cs.CV↗

arXiv2025

A Survey on Diffusion Models for Time Series and Spatio-Temporal Data

Diffusion models have been widely used in time series and spatio-temporal data, enhancing generative, inferential, and downstream capabilities. These models are applied across diverse fields such as healthcare, recommendation, climate, energy, audio, and traffic. By separating applications for time series and spatio-temporal data, we offer a structured perspective on model category, task type, data modality, and practical application domain. This study aims to provide a solid foundation for researchers and practitioners, inspiring future innovations that tackle traditional challenges and foster novel solutions in diffusion model-based data mining tasks and applications. For more detailed information, we have open-sourced a repository at https://github.com/yyysjz1997/Awesome-TimeSeries-SpatioTemporal-Diffusion-Model.

cs.LG↗