Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Refine Now, Query Fast: A Decoupled Refinement Paradigm for Implicit Neural Fields

About

Implicit Neural Representations (INRs) have emerged as promising surrogates for large 3D scientific simulations due to their ability to continuously model spatial and conditional fields, yet they face a critical fidelity-speed dilemma: deep MLPs suffer from high inference cost, while efficient embedding-based models lack sufficient expressiveness. To resolve this, we propose the Decoupled Representation Refinement (DRR) architectural paradigm. DRR leverages a deep refiner network, alongside non-parametric transformations, in a one-time offline process to encode rich representations into a compact and efficient embedding structure. This approach decouples slow neural networks with high representational capacity from the fast inference path. We introduce DRR-Net, a simple network that validates this paradigm, and a novel data augmentation strategy, Variational Pairs (VP) for improving INRs under complex tasks like high-dimensional surrogate modeling. Experiments on several ensemble simulation datasets demonstrate that our approach achieves state-of-the-art fidelity, while being up to 27$\times$ faster at inference than high-fidelity baselines and remaining competitive with the fastest models. The DRR paradigm offers an effective strategy for building powerful and practical neural field surrogates and \rev{INRs in broader applications}, with a minimal compromise between speed and quality.

Tianyu Xiong, Skylar Wurster, Han-Wei Shen• 2026

Related benchmarks

TaskDatasetResultRank
Conditional GeneralizationNyx (test)
Relative L2 Error0.0318
4
Conditional GeneralizationCloverleaf3D (test)
Relative L2 Error0.0981
4
Conditional GeneralizationNyx low-resolution (x2 reduced) (Unseen Fields)
Rel L2 Error0.0111
4
Conditional GeneralizationNyx Trained Fields low-resolution (x2 reduced)
Relative L2 Error0.0115
4
Conditional GeneralizationCloverleaf3D low-resolution (x2 reduced) (Unseen Fields)
Rel L2 Error0.0938
4
Conditional GeneralizationCloverleaf3D low-resolution (x2 reduced) (train)
Relative L2 Error0.0596
4
Super-ResolutionNyx Unseen Fields (test)
Relative L2 Loss0.0431
4
Super-ResolutionNyx Trained Fields (train)
Rel L20.0707
4
Super-ResolutionCloverleaf3D Unseen Fields (test)
Rel L2 Error0.101
4
Super-ResolutionCloverleaf3D Trained Fields (train)
Rel L20.087
4
Showing 10 of 11 rows

Other info

Follow for update