AI Paper Atlas

Research map / 3D vision / Gaussian splatting

Learning Physics-Grounded 4D Dynamics with Neural Gaussian Force Fields

Shiqian Li, Ruihong Shen, Junfeng Ni, Chang Pan, Chi Zhang, Yixin Zhu

ICLR 2026 · 3D vision › Gaussian splatting · approach: Monocular dynamic scene reconstruction · 3 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 soft-body local stress field (deformation modeling) (ablation)NGFF · GSCollision spatialRMSE: 0.110 -> 0.082Improved
add soft-body local stress field (deformation modeling) (ablation)NGFF · GSCollision temporalRMSE: 0.144 -> 0.107Improved
add soft-body local stress field (deformation modeling) (ablation)NGFF · GSCollision compositionalRMSE: 0.131 -> 0.104Improved
add soft-body local stress field (deformation modeling) (ablation)NGFF · GSCollision compositionalcorrelation R: 0.515 -> 0.571Improved
add soft-body local stress field (ablation)NGFF · GSCollisioninference time (s): 0.303 -> 0.363Hurt
learned force fields instead of PointformerNGFF vs Pointformer · GSCollision spatialRMSE: 0.096 -> 0.082Improved
learned force fields instead of PointformerNGFF vs Pointformer · GSCollisioninference time (s): 0.183 -> 0.363Hurt
replace MPM with VLM-estimated params by learned force fieldNGFF vs VLM-MPM · GSCollisioninference time (s): 39.29 -> 0.363Improved
NGFF-V rendering vs Veo3 on photorealismNGFF-V · GSCollision video generation (All, VLM eval)photorealism: 0.41 -> 0.35Hurt
NGFF-V vs Cosmos tuned physical realismNGFF-V · GSCollision video generation (All)physical realism: VLM 0.24 -> 0.30; human 0.59 -> 0.55Mixed
use NGFF-V vs Cosmos for physical realismNGFF-V · GSCollision video generation (All, VLM eval)physical realism: 0.20 -> 0.30Improved

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