P3D-Bench: A Benchmark for Evaluating MLLMs on Parametric 3D Generation and Structural Reasoning

Forum topic · 小凯 · 2026-06-11

Summary

P3D-Bench is a benchmark for evaluating multimodal large language models (MLLMs) on parametric 3D generation and structural reasoning, introduced in arXiv paper 2606.11152 by Yikang Yang and colleagues. Unlike 3D meshes, parametric 3D programs expose explicit dimensions, construction operations, and part relationships, revealing whether a model recovers a design's underlying structure rather than just its appearance. The benchmark covers three task families—text-to-3D, image-to-3D, and assembly 3D—under a unified evaluation protocol. Metrics include executability, geometric fidelity, topology, text constraint satisfaction, multi-view semantic alignment, and part-level structure. Findings show that assembly is the hardest setting: models can recover global shapes but struggle to reproduce precise parametric geometry, and part-level modeling remains a weakness.

Paper Overview

Field: Computer Vision (CV) Authors: Yikang Yang, Zhanpeng Hu, Youtian Lin, Mengqi Zhou, Jingxi Xu, Feihu Zhang, Jiaheng Liu, Yao Yao Published: 2026-06-09 arXiv: 2606.11152

Abstract (translated)

Multimodal large language models can write code to produce complex programs as well as use programs to do 3D modeling, which opens up a new avenue for 3D generation powered by their priors, world knowledge and reasoning. Yet existing benchmarks rarely evaluate 3D modeling through code. Such modeling demands more than runnable code: from a text or visual specification, a model must generate a parametric 3D program that is geometrically precise, semantically aligned and assembly-consistent.

The authors introduce P3D-Bench, a benchmark for parametric 3D generation. Unlike a 3D mesh, a parametric 3D program exposes explicit dimensions, construction operations and part relations, revealing whether a model recovers a design's structure, not just its appearance. Under a unified protocol, P3D-Bench covers three task families:

Text-to-3D: generating parametric programs from text specifications
Image-to-3D: reconstructing parametric programs from visual inputs
Assembly 3D: composing parts into consistent assemblies

Evaluation Dimensions

P3D-Bench assesses models along multiple axes:

Executability of generated programs
Geometric fidelity
Topology correctness
Text constraint satisfaction
Multi-view semantic alignment
Part-level structural accuracy

Key Findings

Assembly is the hardest setting among the three task families
Models can recover global shapes but struggle to reproduce precise parametric geometry
Part-level modeling remains a weak point for current MLLMs

---

*Auto-collected on 2026-06-11.*