Phaedra: Learning High-Fidelity Discrete Tokenization for the Physical Science

• URL: http://arxiv.org/abs/2602.03915v1
• Date: Tue, 03 Feb 2026 17:12:57 GMT
• Title: Phaedra: Learning High-Fidelity Discrete Tokenization for the Physical Science
• Abstract summary: We investigate the accuracy of a suite of image tokenizers to measure the fidelity of PDE properties in both physical and spectral space.<n>Based on the observation that these struggle to capture both fine details and precise magnitudes, we propose Phaedra.<n>We demonstrate that Phaedra consistently improves reconstruction across a range of PDE datasets.
• Score: 16.904886444697347
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Tokens are discrete representations that allow modern deep learning to scale by transforming high-dimensional data into sequences that can be efficiently learned, generated, and generalized to new tasks. These have become foundational for image and video generation and, more recently, physical simulation. As existing tokenizers are designed for the explicit requirements of realistic visual perception of images, it is necessary to ask whether these approaches are optimal for scientific images, which exhibit a large dynamic range and require token embeddings to retain physical and spectral properties. In this work, we investigate the accuracy of a suite of image tokenizers across a range of metrics designed to measure the fidelity of PDE properties in both physical and spectral space. Based on the observation that these struggle to capture both fine details and precise magnitudes, we propose Phaedra, inspired by classical shape-gain quantization and proper orthogonal decomposition. We demonstrate that Phaedra consistently improves reconstruction across a range of PDE datasets. Additionally, our results show strong out-of-distribution generalization capabilities to three tasks of increasing complexity, namely known PDEs with different conditions, unknown PDEs, and real-world Earth observation and weather data.

Related papers

• Image Editing Models are Numerical Solvers [0.0]
  We investigate whether a pretrained generative image-editing model can provide a common interface for numerical simulation.<n>Physical inputs and solutions are rendered as images, while scalar quantities such as material properties, diffusivity, and loading parameters enter through lightweight adapters.<n>Results show that a pretrained image model can represent diverse static and time-dependent physical mappings.
  arXiv  Detail & Related papers  (2026-07-21T07:06:46Z)
• Deep Discrete Encoders: Identifiable Deep Generative Models for Rich Data with Discrete Latent Layers [13.545948734057268]
  We propose interpretable deep generative models for rich data types with discrete latent layers, called Deep Discrete EMs (DDEs)<n>A DDE is a directed graphical model with multiple binary latent layers. Theoretically, we propose transparent identifiability conditions for DDEs, which imply progressively smaller sizes of the latent layers as they go deeper.
  arXiv  Detail & Related papers  (2025-01-02T18:56:23Z)
• Physics Meets Pixels: PDE Models in Image Processing [55.2480439325792]
  Partial Differential Equations (PDEs) have long been recognized as powerful tools for image processing and analysis.<n>We introduce novel physical-based PDE models specifically designed for various image processing tasks.
  arXiv  Detail & Related papers  (2024-12-11T23:11:50Z)
• Geometry Distributions [51.4061133324376]
  We propose a novel geometric data representation that models geometry as distributions. Our approach uses diffusion models with a novel network architecture to learn surface point distributions. We evaluate our representation qualitatively and quantitatively across various object types, demonstrating its effectiveness in achieving high geometric fidelity.
  arXiv  Detail & Related papers  (2024-11-25T04:06:48Z)
• PhyMPGN: Physics-encoded Message Passing Graph Network for spatiotemporal PDE systems [31.006807854698376]
  We propose a new graph learning approach, namely, Physics-encoded Message Passing Graph Network (PhyMPGN)<n>We incorporate a GNN into a numerical integrator to approximate the temporal marching of partialtemporal dynamics for a given PDE system.<n>PhyMPGN is capable of accurately predicting various types of operatortemporal dynamics on coarse unstructured meshes.
  arXiv  Detail & Related papers  (2024-10-02T08:54:18Z)
• Do Graph Neural Networks Work for High Entropy Alloys? [12.002942104379986]
  High-entropy alloys (HEAs) lack chemical long-range order, limiting the applicability of current graph representations. We introduce the LESets machine learning model, an accurate, interpretable GNN for HEA property prediction. We demonstrate the accuracy of LESets in modeling the mechanical properties ofquaternary HEAs.
  arXiv  Detail & Related papers  (2024-08-29T08:20:02Z)
• Invariant Discovery of Features Across Multiple Length Scales: Applications in Microscopy and Autonomous Materials Characterization [3.386918190302773]
  Variational Autoencoders (VAEs) have emerged as powerful tools for identifying underlying factors of variation in image data. We introduce the scale-invariant VAE approach (SI-VAE) based on the progressive training of the VAE with the descriptors sampled at different length scales.
  arXiv  Detail & Related papers  (2024-08-01T01:48:46Z)
• GIC: Gaussian-Informed Continuum for Physical Property Identification and Simulation [60.33467489955188]
  This paper studies the problem of estimating physical properties (system identification) through visual observations. To facilitate geometry-aware guidance in physical property estimation, we introduce a novel hybrid framework. We propose a new dynamic 3D Gaussian framework based on motion factorization to recover the object as 3D Gaussian point sets. In addition to the extracted object surfaces, the Gaussian-informed continuum also enables the rendering of object masks during simulations.
  arXiv  Detail & Related papers  (2024-06-21T07:37:17Z)
• PhyRecon: Physically Plausible Neural Scene Reconstruction [81.73129450090684]
  We introduce PHYRECON, the first approach to leverage both differentiable rendering and differentiable physics simulation to learn implicit surface representations. Central to this design is an efficient transformation between SDF-based implicit representations and explicit surface points. Our results also exhibit superior physical stability in physical simulators, with at least a 40% improvement across all datasets.
  arXiv  Detail & Related papers  (2024-04-25T15:06:58Z)
• GM-NeRF: Learning Generalizable Model-based Neural Radiance Fields from Multi-view Images [79.39247661907397]
  We introduce an effective framework Generalizable Model-based Neural Radiance Fields to synthesize free-viewpoint images. Specifically, we propose a geometry-guided attention mechanism to register the appearance code from multi-view 2D images to a geometry proxy.
  arXiv  Detail & Related papers  (2023-03-24T03:32:02Z)
• Neural Implicit Representations for Physical Parameter Inference from a Single Video [49.766574469284485]
  We propose to combine neural implicit representations for appearance modeling with neural ordinary differential equations (ODEs) for modelling physical phenomena. Our proposed model combines several unique advantages: (i) Contrary to existing approaches that require large training datasets, we are able to identify physical parameters from only a single video. The use of neural implicit representations enables the processing of high-resolution videos and the synthesis of photo-realistic images.
  arXiv  Detail & Related papers  (2022-04-29T11:55:35Z)
• ACID: Action-Conditional Implicit Visual Dynamics for Deformable Object Manipulation [135.10594078615952]
  We introduce ACID, an action-conditional visual dynamics model for volumetric deformable objects. A benchmark contains over 17,000 action trajectories with six types of plush toys and 78 variants. Our model achieves the best performance in geometry, correspondence, and dynamics predictions.
  arXiv  Detail & Related papers  (2022-03-14T04:56:55Z)
• Filtered-CoPhy: Unsupervised Learning of Counterfactual Physics in Pixel Space [43.654464513994164]
  We present a method for learning causal relationships in high-dimensional data (images, videos) Our method does not require the knowledge or supervision of any ground truth positions or other object or scene properties. We introduce a new challenging and carefully designed counterfactual benchmark for predictions in pixel space.
  arXiv  Detail & Related papers  (2022-02-01T12:18:30Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.