Related papers: FlowMixer: A Depth-Agnostic Neural Architecture fo…

Long-term time series forecasting (LTSF) is a critical task in computational intelligence. While Transformer-based models effectively capture long-range dependencies, they often suffer from quadratic complexity and overfitting due to data…

We present the Fourier-Invertible Neural Encoder (FINE), a compact and interpretable architecture for dimension reduction in translation-equivariant datasets. FINE integrates reversible filters and monotonic activation functions with a…

Recently, MLP structures have regained popularity, with MLP-Mixer standing out as a prominent example. In the field of computer vision, MLP-Mixer is noted for its ability to extract data information from both channel and token perspectives,…

Fourier Neural Operators (FNOs) have demonstrated exceptional accuracy in mapping functional spaces by leveraging Fourier transforms to establish a connection with underlying physical principles. However, their opaque inner workings often…

We introduce Neural Flow Maps, a novel simulation method bridging the emerging paradigm of implicit neural representations with fluid simulation based on the theory of flow maps, to achieve state-of-the-art simulation of inviscid fluid…

Time series forecasting plays a critical role in domains such as energy, finance, and healthcare, where accurate predictions inform decision-making under uncertainty. Although Transformer-based models have demonstrated success in sequential…

Fluid-structure interaction is common in engineering and natural systems, where floating-body motion is governed by added mass, drag, and background flows. Modeling these dissipative dynamics is difficult: black-box neural models regress…

For autoregressive modeling of chaotic dynamical systems over long time horizons, the stability of both training and inference is a major challenge in building scientific foundation models. We present a hybrid technique in which an…

Representing 3D surfaces as level sets of continuous functions over $\mathbb{R}^3$ is the common denominator of neural implicit representations, which recently enabled remarkable progress in geometric deep learning and computer vision…

In the last few years, the success of Transformers in computer vision has stimulated the discovery of many alternative models that compete with Transformers, such as the MLP-Mixer. Despite their weak inductive bias, these models have…

Transformer-based architectures are the model of choice for natural language understanding, but they come at a significant cost, as they have quadratic complexity in the input length, require a lot of training data, and can be difficult to…

In this work, we propose an automatic mesh generation algorithm, FlowMesher, which can be used to generate unstructured meshes for mesh domains in any shape with minimum (or even no) user intervention. The approach can generate high-quality…

Forecasting long-term time series in IoT environments remains a significant challenge due to the non-stationary and multi-scale characteristics of sensor signals. Furthermore, error accumulation causes a decrease in forecast quality when…

We report a novel physics-informed neural framework for reconstructing unsteady fluid-structure interactions (FSI) from sparse, single-phase observations of the flow. Our approach combines a modal surface model with coordinate neural…

Monocular depth estimation within the diffusion-denoising paradigm demonstrates impressive generalization ability but suffers from low inference speed. Recent methods adopt a single-step deterministic paradigm to improve inference…

A fundamental problem in robotic perception is matching identical objects or data, with applications such as loop closure detection, place recognition, object tracking, and map fusion. While the problem becomes considerably more challenging…

Standard transformers entangle all computation in a single residual stream, obscuring which components perform which functions. We introduce the Dual-Stream Transformer, which decomposes the residual stream into two functionally distinct…

Fluence map prediction is central to automated radiotherapy planning but remains an ill-posed inverse problem due to the complex relationship between volumetric anatomy and beam-intensity modulation. Convolutional methods in prior work…

Time series forecasting remains a challenging task, particularly in the context of complex multiscale temporal patterns. This study presents LLM-Mixer, a framework that improves forecasting accuracy through the combination of multiscale…

We present FlowCapX, a physics-enhanced framework for flow reconstruction from sparse video inputs, addressing the challenge of jointly optimizing complex physical constraints and sparse observational data over long time horizons. Existing…