3 papers
cs.AI2026
ChronoAgentic: A Code-based Multi-Agent World Simulator for Physically Grounded Simulation Construction
Hongyu Wang, Jingquan Wang, Bocheng Zou +3
Video-based world models generate visually plausible rollouts, but since they infer dynamics in latent states, they enforce no explicit physical constraints: contacts drift, shapes…
cs.LG2026
FNODE: Flow-Matching for data-driven simulation of constrained multibody systems
Hongyu Wang, Jingquan Wang, Dan Negrut
Data-driven modeling of constrained multibody dynamics remains challenged by (i) the training cost of Neural ODEs, which typically require backpropagation through an ODE solver, an…
cs.AI2026
SimBench: A Framework for Evaluating and Diagnosing LLM-Based Digital-Twin Generation for Multi-Physics Simulation
Jingquan Wang, Andrew Negrut, Hongyu Wang +2
We introduce SimBench, a benchmark designed to evaluate the proficiency of simulator-oriented LLMs (S-LLMs) in generating digital twins (DTs) that can be used in simulators for vir…