ExeWorldBench: Benchmarking Code-Based Executable 3D World Modeling
cs.CV
Submitted: 2026-10-10
Updated: 2026-10-10
Code: https://github.com/anthropics/claude-code
Project page: https://helen1p.github.io/ExeWorldBench/ABSTRACT
Terminology
Sources
- Code World Model: Coding Agent as World Brain
- VisRefiner: Learning from Visual Differences for Screenshot-to-Code Generation
- 3DCodeBench: Benchmarking Agentic Procedural 3D Modeling Via Code
- WorldClaw: Agentic 3D Open-World Generation at Scale
- ScreenCoder: Advancing Visual-to-Code Generation for Front-End Automation via Modular Multimodal Agents
- SimWorld Studio: Automatic Environment Generation with Evolving Coding Agent for Embodied Agent Learning
- MPMWorlds: Material-Point-Method Simulations for Inferring and Extrapolating Physical Dynamics
- VisPhyWorld: Probing Physical Reasoning via Code-Driven Video Reconstruction
- WorldCoder-Bench: Benchmarking Physically Grounded 3D World Synthesis
- Nova3D: Code-Native Generation of Programmable 3D Assets
- PhysVista: Benchmarking Physical Intelligence in VLMs via a Perception-Reasoning-Assessment Loop
- Coding Agent Is Good As World Simulator
- Training and Benchmarking Code Generation for Physics-Inspired Animations
- PhysCodeBench: Benchmarking Physics-Aware Symbolic Simulation of 3D Scenes via Self-Corrective Multi-Agent Refinement
- PhysMind: From Video to Executable Worlds for Training-Free Physical Reasoning
- BenchCAD: A Comprehensive, Industry-Standard Benchmark for Programmatic CAD
- From Visual Widgets to UI Code: Efficient Tool-Grounded Generation
- Code2Worlds: Empowering Coding LLMs for 4D World Generation
Related papers
- Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
- AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
- Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
- MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
- TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
- A Survey on Efficient Vision-Language-Action Models