OMNIFLOW: A Physics-Grounded Multimodal Agent for Generalized Scientific Reasoning

TL;DR

OMNIFLOW is a neuro-symbolic framework that enhances multimodal LLMs with physical law grounding, enabling better scientific reasoning and generalization across diverse physics-based tasks without domain-specific fine-tuning.

Contribution

It introduces a Semantic-Symbolic Alignment mechanism and a Physics-Guided Chain-of-Thought workflow to incorporate physical constraints into LLM reasoning.

Findings

Outperforms traditional deep learning models in zero-shot and few-shot scenarios.

Provides transparent, physically consistent reasoning reports.

Demonstrates strong generalization across turbulence, Navier-Stokes, and weather forecasting tasks.

Abstract

Large Language Models (LLMs) have demonstrated exceptional logical reasoning capabilities but frequently struggle with the continuous spatiotemporal dynamics governed by Partial Differential Equations (PDEs), often resulting in non-physical hallucinations. Existing approaches typically resort to costly, domain-specific fine-tuning, which severely limits cross-domain generalization and interpretability. To bridge this gap, we propose OMNIFLOW, a neuro-symbolic architecture designed to ground frozen multimodal LLMs in fundamental physical laws without requiring domain-specific parameter updates. OMNIFLOW introduces a novel \textit{Semantic-Symbolic Alignment} mechanism that projects high-dimensional flow tensors into topological linguistic descriptors, enabling the model to perceive physical structures rather than raw pixel values. Furthermore, we construct a Physics-Guided Chain-of-Thought (PG-CoT) workflow that orchestrates reasoning through dynamic constraint injection (e.g., mass conservation) and iterative reflexive verification. We evaluate OMNIFLOW on a comprehensive benchmark spanning microscopic turbulence, theoretical Navier-Stokes equations, and macroscopic global weather forecasting. Empirical results demonstrate that OMNIFLOW significantly outperforms traditional deep learning baselines in zero-shot generalization and few-shot adaptation tasks. Crucially, it offers transparent, physically consistent reasoning reports, marking a paradigm shift from black-box fitting to interpretable scientific reasoning.