FLUID-LLM: Learning Computational Fluid Dynamics with Spatiotemporal-aware Large Language Models

Max Zhu, Adrián Bazaga, Pietro Liò
6/6/2024
cs.LGcs.AIcs.CL

No official code found. Best implementation: AdrianBZG/FLUID-LLM (5 stars, MIT, last commit 2024-12-17).

Abstract

Learning computational fluid dynamics (CFD) traditionally relies on computationally intensive simulations of the Navier-Stokes equations. Recently, large language models (LLMs) have shown remarkable pattern recognition and reasoning abilities in natural language processing (NLP) and computer vision (CV). However, these models struggle with the complex geometries inherent in fluid dynamics. We introduce FLUID-LLM, a novel framework combining pre-trained LLMs with spatiotemporal-aware encoding to predict unsteady fluid dynamics. Our approach leverages the temporal autoregressive abilities of LLMs alongside spatial-aware layers, bridging the gap between previous CFD prediction methods. Evaluations on standard benchmarks reveal significant performance improvements across various fluid datasets. Our results demonstrate that FLUID-LLM effectively integrates spatiotemporal information into pre-trained LLMs, enhancing CFD task performance.

View on arXivView PDF

Code Implementations(1)

AdrianBZG/FLUID-LLMCommunity74%
MIT

FLUID-LLM: Learning Computational Fluid Dynamics with Spatiotemporal-aware Large Language Models

52Shell, PythonFeb 7, 20241 years agoMIT
deep-learningfluid-dynamicslanguage-modelspatiotemporal

Similar Papers

CFDLLMBench: A Benchmark Suite for Evaluating Large Language Models in Computational Fluid Dynamics

85% match2025

Hybrid machine learning models based on physical patterns to accelerate CFD simulations: a short guide on autoregressive models

84% match2025

Fine-tuning a Large Language Model for Automating Computational Fluid Dynamics Simulations

84% match1 repo472025

Recent Advances on Machine Learning for Computational Fluid Dynamics: A Survey

83% match1 repo3652024

Data-Efficient Inference of Neural Fluid Fields via SciML Foundation Model

83% match1 repo62024

Spatio-Temporal Fluid Dynamics Modeling via Physical-Awareness and Parameter Diffusion Guidance

81% match2024

Cite this paper

@article{zhu2024fluidllm,
  title  = {FLUID-LLM: Learning Computational Fluid Dynamics with Spatiotemporal-aware Large Language Models},
  author = {Max Zhu and Adrián Bazaga and Pietro Liò},
  year   = {2024},
  eprint = {2406.04501},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2406.04501v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2406.04501.svg)](https://codeofpaper.com/paper/2406.04501)

Discussion