We present NFF, a modeling framework built on NODE that learns interpretable force field representations which can be efficiently integrated through an ODE solver to predict object trajectories. Unlike existing approaches that rely on high-dimensional latent spaces, NFF captures fundamental physical concepts such as gravity, support, and collision in an interpretable manner. Experiments on two challenging physical reasoning tasks demonstrate that NFF, trained with only a few examples, achieves strong generalization to unseen scenarios. This physics-grounded representation enables efficient forward-backward planning and rapid adaptation through interactive refinement.
Neural Force Field: Few shot learning of generalized physical reasoning
Shiqian Li
, Ruihong Shen
, Yaoyu Tao
, Chi Zhang
, Yixin Zhu
Arxiv
Github
ICLR