AI Paper Atlas

Research map / Probabilistic and scientific modeling / Dynamical systems

Contact-Aware Neural Dynamics

Changwei Jing, Jai Krishna Bandi, Jianglong Ye, Yan Duan, Pieter Abbeel, Xiaolong Wang, Sha Yi

CVPR 2026 · Probabilistic and scientific modeling › Dynamical systems · approach: Physics-informed dynamics learning · 5 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add binary contact prediction and conditioning to Diffusion-UNet (ablation)Diffusion-UNet · single object, real fine-tuneADD-S AUC (%): 82.45 -> 88.23Improved
add binary contact prediction and conditioning to Diffusion-UNet (ablation)Diffusion-UNet · single object, real fine-tuneMSE: 0.0091 -> 0.0082Improved
add binary contact prediction and conditioning to Diffusion-UNet (ablation)Diffusion-UNet · multiple objects, real fine-tuneADD-S AUC (%): 75.82 -> 79.12Improved
add binary contact prediction and conditioning to Diffusion-UNet (ablation)Diffusion-UNet · single object, sim dataADD-S AUC (%): 69.12 -> 68.12Hurt
add binary contact prediction and conditioning to Diffusion-UNet (ablation)Diffusion-UNet · single object, real dataADD-S AUC (%): 80.03 -> 81.34Improved
add binary contact prediction and conditioning to Diffusion-UNet (ablation)Diffusion-UNet · multiple objects, real dataMSE: 0.0083 -> 0.0075Improved
replace MLP with Diffusion-UNet dynamics head (ablation)MLP · single object, real dataADD-S AUC (%): 78.12 -> 80.03Improved
replace plain UNet with diffusion UNet (ablation)UNet · single object, real dataADD-S AUC (%): 68.45 -> 80.03Improved
fine-tune on real data after simulation pretrainingDiffusion-UNet w/ Contact · single objectADD-S AUC (%): 81.34 -> 88.23Improved
fine-tune on real data after simulation pretrainingMLP · single objectADD-S AUC (%): 78.12 -> 77.43Hurt
pretrain on simulation then fine-tune on real data instead of real-only trainingcontact-aware dynamics · single-object task successsuccess rate (%): 52.6 -> 73.7Improved
pretrain on simulation then fine-tune on real data instead of real-only trainingcontact-aware dynamics · multi-object task successsuccess rate (%): 47.1 -> 64.7Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy