Frontier3DBench: Benchmarking Coding Agents for 3D Creation across Objects, Scenes, and Interactive Open Worlds
cs.CV
Submitted: 2026-11-06
Updated: 2026-11-06
Project page: https://lab.einsia.ai/frontier3dbench
Terminology
Sources
- Vision-as-Inverse-Graphics Agent via Interleaved Multimodal Reasoning
- Code2Games: Enabling Coding Agents for Gaming World Generation
- BlenderWorld: Benchmarking Visual Agents for 3D Creation from Real-World References
- Code4Scene: Benchmarking Coding Agents for Constructing and Editing 3D Scenes
- ExeWorldBench: Benchmarking Code-Based Executable 3D World Modeling
- From Code to Playable Systems: CEGame-Bench for Long-Horizon Game Creation Agents
- BenchCAD: A Comprehensive, Industry-Standard Benchmark for Programmatic CAD
- P3D-Bench: Benchmarking MLLMs for Parametric 3D Generation and Structural Reasoning
- RealCADBench: Benchmarking Parametric CAD Modeling from Industrial Design Intents
- IndustrialCAD-1400: Benchmarking Detailed Industrial Assemblies from Text
- 3DCodeBench: Benchmarking Agentic Procedural 3D Modeling Via Code
- Benchmarking PhD-Level Coding in 3D Geometric Computer Vision
- MUSE: Agentic 3D Scene Authoring via Memory-Grounded Incremental Requirement Satisfaction
- VoxelCodeBench: Benchmarking 3D World Modeling Through Code Generation
- SceneActBench: Can Agents Act on the 3D Scenes They See?
- NeoWorld-3: From Pixels to Code-Defined Physical Worlds
- SceneCode: Executable World Programs for Editable Indoor Scenes with Articulated Objects
- EmbodiedGen V2: An Agentic, Simulation-Ready 3D World Engine for Embodied AI
- Code2Worlds-2: Physically Plausible Agent Harness for Embodied World Simulation
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs