500 citations · 2k across the 65 of their papers we have counts for
70 papers
Transolver-: Joint Spectral-Physical Subspace Modeling for Neural PDE Solving
Haonan Shangguan, Hang Zhou, Haixu Wu +3
Neural solvers offer efficient surrogates for numerical simulation of partial differential equations (PDEs). For time-dependent problems, strong one-step accuracy does not necessar…
NowcastDiT: Diffusion Transformers are Effective Precipitation Nowcasters
Haoran Xu, Xingzhuo Guo, Yuchen Zhang +3
Precipitation nowcasting demands accurate short-term forecasts under strong spatiotemporal variability. Diffusion models are well suited to modeling complex precipitation distribut…
Harnessing AI for Inverse Partial Differential Equation Problems: Past, Present, and Prospects
Zhentao Tan, Yuze Hao, Boyi Zou +3
Solving inverse partial differential equation (PDE) problems is a fundamental topic in scientific research due to its broad significance across a wide range of real-world applicati…
Timer-S1: A Billion-Scale Time Series Foundation Model with Serial Scaling
Yong Liu, Xingjian Su, Shiyu Wang +7
We introduce Timer-S1, a strong Mixture-of-Experts (MoE) time series foundation model with 8.3B total parameters, 0.75B activated parameters for each token, and a context length of…
Regression Models Meet Foundation Models: A Hybrid-AI Approach to Practical Electricity Price Forecasting
Yunzhong Qiu, Binzhu Li, Hao Wei +5
Electricity market prices exhibit extreme volatility, nonlinearity, and non-stationarity, making accurate forecasting a significant challenge. While cutting-edge time series founda…
BTTackler: A Diagnosis-based Framework for Efficient Deep Learning Hyperparameter Optimization
Zhongyi Pei, Zhiyao Cen, Yipeng Huang +4
Hyperparameter optimization (HPO) is known to be costly in deep learning, especially when leveraging automated approaches. Most of the existing automated HPO methods are accuracy-b…
DiTS: Multimodal Diffusion Transformers Are Time Series Forecasters
Haoran Zhang, Haixuan Liu, Yong Liu +4
While generative modeling facilitates probabilistic time series forecasting, incorporating heterogeneous exogenous information remains challenging. Diffusion Transformers (DiT) pro…
GeoPT: Scaling Physics Simulation via Lifted Geometric Pre-Training
Haixu Wu, Minghao Guo, Zongyi Li +4
Neural simulators promise efficient surrogates for physics simulation, but scaling them is bottlenecked by the prohibitive cost of generating high-fidelity training data. Pre-train…
It's TIME: Towards the Next Generation of Time Series Forecasting Benchmarks
Zhongzheng Qiao, Sheng Pan, Anni Wang +7
Time series foundation models (TSFMs) are revolutionizing the forecasting landscape from specific dataset modeling to generalizable task evaluation. However, we contend that existi…
Brep2Shape: Boundary and Shape Representation Alignment via Self-Supervised Transformers
Yuanxu Sun, Yuezhou Ma, Haixu Wu +4
Boundary representation (B-rep) is the industry standard for computer-aided design (CAD). While deep learning shows promise in processing B-rep models, existing methods suffer from…
Transolver-3: Scaling Up Transformer Solvers to Industrial-Scale Geometries
Hang Zhou, Haixu Wu, Haonan Shangguan +4
Deep learning has emerged as a transformative tool for the neural surrogate modeling of partial differential equations (PDEs), known as neural PDE solvers. However, scaling these s…
CoRA: Covariate-Aware Adaptation of Time Series Foundation Models
Guo Qin, Zhi Chen, Yong Liu +5
Time Series Foundation Models (TSFMs) have shown significant impact through their model capacity, scalability, and zero-shot generalization. However, due to the heterogeneity of in…
Mitigating the Noise Shift for Denoising Generative Models via Noise Awareness Guidance
Jincheng Zhong, Boyuan Jiang, Xin Tao +3
Existing denoising generative models rely on solving discretized reverse-time SDEs or ODEs. In this paper, we identify a long-overlooked yet pervasive issue in this family of model…
Exploring Accuracy Law for Deep Time Series Forecasters: An Empirical Study
Yuxuan Wang, Haixu Wu, Yuezhou Ma +8
Deep time series forecasting has emerged as a rapidly growing field in recent years. Despite the exponential growth of community interests, progress on standard benchmarks is often…
PhySense: Sensor Placement Optimization for Accurate Physics Sensing
Yuezhou Ma, Haixu Wu, Hang Zhou +3
Physics sensing plays a central role in many scientific and engineering domains, which inherently involves two coupled tasks: reconstructing dense physical fields from sparse obser…
FlashBias: Fast Computation of Attention with Bias
Haixu Wu, Minghao Guo, Yuezhou Ma +4
Attention with bias, which extends standard attention by introducing prior knowledge as an additive bias matrix to the query-key scores, has been widely deployed in vision, languag…
Domain Guidance: A Simple Transfer Approach for a Pre-trained Diffusion Model
Jincheng Zhong, Xiangcheng Zhang, Jianmin Wang +1
Recent advancements in diffusion models have revolutionized generative modeling. However, the impressive and vivid outputs they produce often come at the cost of significant model…
Dynamical Diffusion: Learning Temporal Dynamics with Diffusion Models
Xingzhuo Guo, Yu Zhang, Baixu Chen +3
Diffusion models have emerged as powerful generative frameworks by progressively adding noise to data through a forward process and then reversing this process to generate realisti…
TimesBERT: A BERT-Style Foundation Model for Time Series Understanding
Haoran Zhang, Yong Liu, Yunzhong Qiu +4
Time series analysis is crucial in diverse scenarios. Beyond forecasting, considerable real-world tasks are categorized into classification, imputation, and anomaly detection, unde…
Transolver++: An Accurate Neural Solver for PDEs on Million-Scale Geometries
Huakun Luo, Haixu Wu, Hang Zhou +4
Although deep models have been widely explored in solving partial differential equations (PDEs), previous works are primarily limited to data only with up to tens of thousands of m…
Sundial: A Family of Highly Capable Time Series Foundation Models
Yong Liu, Guo Qin, Zhiyuan Shi +5
We introduce Sundial, a family of native, flexible, and scalable time series foundation models. To predict the next-patch's distribution, we propose a TimeFlow Loss based on flow-m…
ProPINN: Demystifying Propagation Failures in Physics-Informed Neural Networks
Yuezhou Ma, Haixu Wu, Hang Zhou +3
Physics-informed neural networks (PINNs) have earned high expectations in solving partial differential equations (PDEs), but their optimization usually faces thorny challenges due…
Metadata Matters for Time Series: Informative Forecasting with Transformers
Jiaxiang Dong, Haixu Wu, Yuxuan Wang +3
Time series forecasting is prevalent in extensive real-world applications, such as financial analysis and energy planning. Previous studies primarily focus on time series modality,…
Timer-XL: Long-Context Transformers for Unified Time Series Forecasting
Yong Liu, Guo Qin, Xiangdong Huang +2
We present Timer-XL, a causal Transformer for unified time series forecasting. To uniformly predict multidimensional time series, we generalize next token prediction, predominantly…