Towards Physically Grounded Articulated 3D Assets: A Multi-Agent Pipeline and Benchmark
Artisan uses multi-agent simulation feedback to generate articulated 3D assets with collision-free motion, stability, and accurate physical properties, achieving the highest stability and lowest collision rates in benchmarks.
Published Oct 9, 2026▲ 3 on Hugging FaceCodearXiv ↗
Only vote on papers you've read. Sign in to vote.
Abstract
In this paper, we study physically grounded articulated 3D asset generation---moving the goal from reproducing how objects look to capturing how they move and function. Inspired by the literature on procedural and agentic generation, we present Artisan, a system that extends the latest method Articraft by introducing a multi-agent pipeline with physical simulation feedback, enabling comprehensive attributes beyond geometry, appearance, and physically plausible articulation, such as collision-free motion, stability under gravity, and accurate mass and material properties. Furthermore, we construct a unified benchmark spanning four key aspects: spatial fidelity, physical plausibility, downstream task performance, and overall quality. The benchmark encompasses 100 reference assets from ArtVIP and LightWheel, 50 stability prompts, and 50 ABO objects with reference masses, with protocols evaluating various metrics. Through extensive evaluation of more than 10 representative models, Artisan achieves competitive spatial fidelity while attaining the lowest collision rates and highest stability rates among the evaluated methods. Together, our system and benchmark establish insights for advancing 3D generation towards physically plausible, functionally usable assets.