# Reduced-order Neural Modeling with Differentiable Simulation for High-Detail Tactile Perception

arXiv: https://arxiv.org/abs/2605.05053 · HTML: https://mlsanity.com/paper/2605.05053/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 27 of its 41 references that are in the map.
- First posted on arXiv 6 May 2026.

## Builds on (most cited first, 27 of 27)

- [PlasticineLab: A Soft-Body Manipulation Benchmark with Differentiable Physics](https://mlsanity.com/paper/2104.03311/) (2021)
- [TACTO: A Fast, Flexible, and Open-source Simulator for High-Resolution Vision-based Tactile Sensors](https://mlsanity.com/paper/2012.08456/) (2020)
- [DiffPD: Differentiable Projective Dynamics](https://mlsanity.com/paper/2101.05917/) (2021)
- [GelSlim: A High-Resolution, Compact, Robust, and Calibrated Tactile-sensing Finger](https://mlsanity.com/paper/1803.00628/) (2018)
- [Taxim: An Example-based Simulation Model for GelSight Tactile Sensors](https://mlsanity.com/paper/2109.04027/) (2021)
- [GelSight Wedge: Measuring High-Resolution 3D Contact Geometry with a Compact Robot Finger](https://mlsanity.com/paper/2106.08851/) (2021)
- [High-order Differentiable Autoencoder for Nonlinear Model Reduction](https://mlsanity.com/paper/2102.11026/) (2021)
- [Sim-to-Real for Robotic Tactile Sensing via Physics-Based Simulation and Learned Latent Projections](https://mlsanity.com/paper/2103.16747/) (2021)
- [TacSL: A Library for Visuotactile Sensor Simulation and Learning](https://mlsanity.com/paper/2408.06506/) (2024)
- [DIFFTACTILE: A Physics-based Differentiable Tactile Simulator for Contact-rich Robotic Manipulation](https://mlsanity.com/paper/2403.08716/) (2024)
- [Neural Stress Fields for Reduced-order Elastoplasticity and Fracture](https://mlsanity.com/paper/2310.17790/) (2023)
- [Tactile Sim-to-Real Policy Transfer via Real-to-Sim Image Translation](https://mlsanity.com/paper/2106.08796/) (2021)
- [Data-Free Learning of Reduced-Order Kinematics](https://mlsanity.com/paper/2305.03846/) (2023)
- [Dense Tactile Force Distribution Estimation using GelSlim and inverse FEM](https://mlsanity.com/paper/1810.04621/) (2018)
- [ManiFeel: Benchmarking and Understanding Visuotactile Manipulation Policy Learning](https://mlsanity.com/paper/2505.18472/) (2025)
- [Model reduction for the material point method via an implicit neural representation of the deformation map](https://mlsanity.com/paper/2109.12390/) (2021)
- [DigiTac: A DIGIT-TacTip Hybrid Tactile Sensor for Comparing Low-Cost High-Resolution Robot Touch](https://mlsanity.com/paper/2206.13657/) (2022)
- [Tacchi: A Pluggable and Low Computational Cost Elastomer Deformation Simulator for Optical Tactile Sensors](https://mlsanity.com/paper/2301.08343/) (2023)
- [Amortized Synthesis of Constrained Configurations Using a Differentiable Surrogate](https://mlsanity.com/paper/2106.09019/) (2021)
- [TacIPC: Intersection- and Inversion-free FEM-based Elastomer Simulation For Optical Tactile Sensors](https://mlsanity.com/paper/2311.05843/) (2023)
- [Accelerate Neural Subspace-Based Reduced-Order Solver of Deformable Simulation by Lipschitz Optimization](https://mlsanity.com/paper/2409.03807/) (2024)
- [Soft Robot Optimal Control Via Reduced Order Finite Element Models](https://mlsanity.com/paper/2011.02092/) (2020)
- [Tacchi 2.0: A Low Computational Cost and Comprehensive Dynamic Contact Simulator for Vision-based Tactile Sensors](https://mlsanity.com/paper/2503.09100/) (2025)
- [Spring-IMU Fusion Based Proprioception for Feedback Control of Soft Manipulators](https://mlsanity.com/paper/2309.14279/) (2023)
- [Learning Reduced-Order Soft Robot Controller](https://mlsanity.com/paper/2311.01720/) (2023)
- [Proprioceptive State Estimation for Amphibious Tactile Sensing](https://mlsanity.com/paper/2312.09863/) (2023)
- [Tactile Displays Driven by Projected Light](https://mlsanity.com/paper/2410.05494/) (2024)

## Abstract (arXiv)

Tactile perception is key to dexterous manipulation, yet simulating high-resolution elastomer deformation remains computationally prohibitive. Finite element methods (FEM) deliver high fidelity but demand costly remeshing, while Material Point Methods (MPM) suffer from heavy particle-memory tradeoffs. We propose a reduced-order neural simulation framework that couples coarse-grained MPM dynamics with an implicit neural decoder to reconstruct sub-particle tactile details from compact latent states. The framework learns a continuous deformation manifold from paired high- and low-resolution simulations, enabling physically consistent, differentiable inference. Compared to the TacIPC, our method achieves over 65% faster simulation and 40% lower memory usage, while maintaining better geometric fidelity. In tactile rendering and 3D surface reconstruction, our methods further improve accuracy by 25% and produce realistic depth images and surface mesh within a faster inference speed. These results demonstrate that the proposed reduced-order neural model enables high-detail, physically grounded tactile simulation with substantial efficiency gains for robotic interaction and optimization.