ScratchWorld: Evaluating If World Models Compute Executable Consequences
cs.SE
Submitted: 2026-06-30
Updated: 2026-10-28
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- VideoPhy: Evaluating Physical Commonsense for Video Generation
- Mastering Diverse Domains through World Models
- Raven: Rethinking Automated Assessment for Scratch Programs via Video-Grounded Evaluation
- VisPhyWorld: Probing Physical Reasoning via Code-Driven Video Reconstruction
- IntPhys: A Framework and Benchmark for Visual Intuitive Physics Reasoning
- ScratchEval : A Multimodal Evaluation Framework for LLMs in Block-Based Programming
- ViScratch: Using Large Language Models and Gameplay Videos for Automated Feedback in Scratch
- Stitch: Step-by-step LLM Guided Tutoring for Scratch
- EcoScratch: Cost-Effective Multimodal Repair for Scratch Using Execution Feedback
- ScratchLens: Lens-Parametric Behavioral Equivalence for Scratch Programs
- World-in-World: World Models in a Closed-Loop World
- See, Plan, Snap: Evaluating Multimodal GUI Agents in Scratch
Related papers
- ArkEval: Benchmarking and Evaluating Automated CodeRepair for ArkTS
- TestPrism: Rethinking Test Evaluation Beyond a Single Reference
- Mining Workflow Graphs for Black-Box Boundary Testing of Conversational LLM Agents
- Falsification-Based Verification of LLM-Generated Optimization Models: Sound Test Batteries and Their Detection Limits
- GitSkills: A Dataset of Agent Skills on GitHub
- For Coding Agents, the Repository Matters