Attention-Guided Flow-Matching for Sparse 3D Geological Generation

2026-04-07

AI Summary

Constructing high-resolution 3D geological models from sparse one-dimensional borehole and two-dimensional surface data entails recovering complex nonlinear structures, posing a highly ill-posed inverse problem. This work proposes 3D-GeoFlow, a novel framework that introduces attention-guided continuous flow matching into sparse multimodal geological modeling for the first time. By leveraging vector field regression and optimizing optimal transport paths, the method circumvents the representational collapse commonly observed in diffusion models under sparse conditions. It incorporates a 3D attention-gated mechanism to dynamically propagate local borehole features, ensuring macroscopic structural consistency. Evaluated on a large-scale dataset comprising 2,200 procedurally generated cases, 3D-GeoFlow significantly outperforms conventional interpolation and diffusion-based baselines, demonstrating exceptional generalization and generation quality in out-of-distribution scenarios.

Technology Category

• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Web Mining and Content Analysis: Models for Web evolution

Abstract

Constructing high-resolution 3D geological models from sparse 1D borehole and 2D surface data is a highly ill-posed inverse problem. Traditional heuristic and implicit modeling methods fundamentally fail to capture non-linear topological discontinuities under extreme sparsity, often yielding unrealistic artifacts. Furthermore, while deep generative architectures like Diffusion Models have revolutionized continuous domains, they suffer from severe representation collapse when conditioned on sparse categorical grids. To bridge this gap, we propose 3D-GeoFlow, the first Attention-Guided Continuous Flow Matching framework tailored for sparse multimodal geological modeling. By reformulating discrete categorical generation as a simulation-free, continuous vector field regression optimized via Mean Squared Error, our model establishes stable, deterministic optimal transport paths. Crucially, we integrate 3D Attention Gates to dynamically propagate localized borehole features across the volumetric latent space, ensuring macroscopic structural coherence. To validate our framework, we curated a large-scale multimodal dataset comprising 2,200 procedurally generated 3D geological cases. Extensive out-of-distribution (OOD) evaluations demonstrate that 3D-GeoFlow achieves a paradigm shift, significantly outperforming heuristic interpolations and standard diffusion baselines.

Problem

• sparse 3D geological modeling
• ill-posed inverse problem
• topological discontinuities
• categorical grid generation
• multimodal geological data

Innovation

• Attention-Guided Flow Matching
• Sparse 3D Geological Modeling
• Continuous Vector Field Regression
• 3D Attention Gates
• Optimal Transport