Overview

Differentiable physics studies how to compute gradients through simulations of systems such as rigid-body contact, deformable materials, fluids, and chemical processes. Common approaches differentiate numerical solvers with automatic differentiation, adjoints, or implicit differentiation, and combine FEM, MPM, or PDE solvers with learned models; smoothing is often used for contacts and discrete events. These gradients support inverse problems, where observations are used to infer physical parameters, as well as trajectory optimization and control in settings such as robot manipulation and fluid engineering. GPU acceleration is making larger simulations practical, but gradients can still be unreliable around contact and other discontinuities, and speed must be balanced against physical fidelity and numerical stability.

Latest papers

CardsList

73 papers5 papers in the last four weeks, down 71% on the four weeks before. 0.0% of all new papers.

  1. Magflow: Differentiable Simulation of Industrial-Grade 6D Magnetic Levitation

    Nov 17, 2026Philipp Hartmann, Jannick Stranghöner, Klaus NeumannFeedback ControlReal-Time Robot Control

  2. tmol: a GPU-accelerated implementation of the Rosetta energy function for machine learning

    Nov 16, 2026Andrew Leaver-Fay, Kieran Didi, Jeff Flatten +12Differentiable PhysicsProtein Design

  3. Sample-Efficient Control of Magnetoconvective Flows via Differentiable Simulation

    Nov 13, 2026Jannis Becktepe, Dmitry Krasnov, Nils Thuerey +2MagnetohydrodynamicsDifferentiable Physics

  4. JAX-WAVE: A Differentiable Physics-Supervised Framework for Inverse Wave Design in Heterogeneous Media

    Nov 10, 2026Honghui Du, Binyao Guo, Yushu PanInverse DesignPDE Inverse Problems

  5. Same Physics, Different Gradients: A State-of-the-Art Report on Differentiable Physics Simulation

    Nov 9, 2026Lei Shu, Zizhou Huang, Yunuo Chen +15adjoint methodDifferentiable Physics

  6. DiffContact: Differentiable Contact Dynamics for Policy Learning and Contact-Implicit Trajectory Optimization

    Oct 26, 2026Gijeong Kim, Seungbin Oh, Dongyun Kang +1Contact-Implicit Trajectory OptimizationDifferentiable Physics

  7. CQDC-Warp: GPU-Accelerated Differentiable Quasi-Dynamic Simulation for Contact-Rich Manipulation

    Oct 15, 2026Yanchen Ding, Jiayi Chen, Jianfeng Xu +3Contact-Rich Robotic ManipulationTrajectory Optimization

  8. DiffGIPC: Differentiable Mixed-Coordinate Contact Mechanics on GPU

    Oct 13, 2026Kemeng Huang, Bowen Ren, Xinyu Lu +2Differentiable PhysicsInverse Problems

  9. TorchSpin: Differentiable Electron Paramagnetic Resonance Simulation in PyTorch

    Oct 12, 2026Alec H. FollmerMagnetic Resonance SpectroscopyDifferentiable Physics

  10. Observation-Referenced Tissue Deformation Forecasting with Tool-Driven and Residual Dynamics

    Oct 9, 2026Tom Olesch, Nan Xiao, Xiongye Xiao +2Deformable Object SimulationSurgical Robotics

  11. A Physics-Informed Collision Learning Framework for Collaborative Robot Motion Generation

    Oct 8, 2026Chen Cai, Steven LiuRobot Collision AvoidanceRobot Motion Generation

  12. UWB Meets Crazyflow: Simulating Degraded Feedback at Scale for Aerial Robotics

    Oct 6, 2026Martin Schuck, Marcel P. Rath, Yufei Hua +3Aerial RoboticsRobotics Simulation

  13. DexForge: High-Fidelity Physics-Informed Dexterous Retargeting

    Oct 5, 2026Meizhong Wang, Kun Cao, Ruiqi Ni +2Contact-Rich Robotic ManipulationMotion Retargeting

  14. An End-to-End Framework for Modelling Pneumatic Soft Robots Based on Differentiable Finite Element Methods

    Oct 3, 2026Shaohong Zhong, Yao Yao, Perla Maiolino +1Soft RoboticsRobot Dynamics Identification

  15. On the Numerical Reliability of Differentiable Physics-Based Optimization for Robotic Material Manipulation

    Sep 28, 2026Xintong Yang, Minglun Wei, Yu-Kun Lai +1Robot Dynamics IdentificationMaterial Point Method

  16. Differentiable Dynamics for Autonomous Micro-Mobility Navigation

    Sep 25, 2026Grace Cai, Joey Lee, Nithin Parepally +2Vehicle DynamicsDifferentiable Physics

  17. Bundled Contact Gradients: Stabilizing Differentiable Simulation for Deployable Dynamic Tasks

    Sep 25, 2026Dyuman Aditya, Jin Cheng, Clemens Schwarke +4Robotics SimulationDifferentiable Physics

  18. CraftSPH: A composable framework for differentiable smoothed particle hydrodynamics implemented in PyTorch

    Sep 24, 2026Gen Matono, Shujiro Fujioka, Mayuko NishioComputational Fluid DynamicsDifferentiable Physics

  19. Adapting Rigid-Body Dynamics Derivatives for Constraint Embedding Closed-Chain Models

    Sep 17, 2026Daniel J. Volpi, Patrick M. WensingMultibody DynamicsDifferentiable Physics

  20. Benchmarking Optimizers to Solve Inverse Problems with Differentiable Physics Simulators

    Sep 12, 2026Xiang Chen, Huanhuan XiaDifferentiable OptimizationInverse Problems

  21. Ostrich: Taking Large Strides Through Stiff Contact in Differentiable Dynamics

    Sep 8, 2026Aleš Kučera, Karel ZimmermannTrajectory OptimizationContact-Rich Manipulation

  22. Differentiable Hybrid Modelling for Learning and Optimising Chemical Transport Processes from Experimental Data

    Sep 3, 2026Arthur Jessop, Mohammed Alsubeihi, Ben Moseley +1Differentiable PhysicsDifferentiable Optimization

  23. From Numerical Simulators of PDEs to Neural Emulators and Back

    Aug 25, 2026Felix KoehlerPDE Surrogate ModelingPDE Solving

  24. Differentiate the Solver, Not the Equation: Reverse-Sweep Adjoints for Block Implicit Simulation

    Aug 9, 2026Lei Shu, Ying Jiang, Kui Wu +3Automatic DifferentiationDifferentiable Programming