PhysLDM: Latent Diffusion Delivers High-Fidelity Deformable Physics Simulation with Zero-Shot Generalization
mmlab-ntu · hf · 2026-10-21
NTU MMLab introduced PhysLDM, a spatiotemporal latent diffusion framework for high-fidelity deformable body simulation.
- Combines a holistic VAE with diffusion to model chaotic dynamics, accurately reconstructing complex deformations like a mug being squashed and rebounding
- Enables zero-shot generalization to diverse unseen objects and scenes
- Demonstrated applications include optimizing material properties to drive objects toward target trajectories or frames
Related event: NTU Releases PhysLDM for High-Fidelity Deformable Physics Simulation(2 posts)→
More from Embodied
- Humanoid robot bends over a car to inspect a person lying underneath — CyberRobooo · 2026-10-21
- Hugging Face Launches SimReady Central: Open Sim-Ready 3D Assets for Robotics — Hugging Face Blog · 2026-10-21
- NVIDIA rolls out local AI lineup at GTC Berlin: DGX Spark 64GB ships, 8-unit clusters run frontier models — NVIDIA Blog · 2026-10-21
- Musk: Neuralink's Blindsight implant writes to visual cortex, first low-res version coming soon — XFreeze · 2026-10-21
- NVIDIA GTC Berlin: Mercedes, Volvo adopt Hyperion as Halos SAFE and Alpamayo 2 Edge debut — nvidia · 2026-10-21
- NVIDIA expands Isaac robotics platform with Agile ONE SN reference humanoid — nvidia · 2026-10-21