ScratchWorld Benchmark
Explore score history ↗ScratchWorld is an offline diagnostic benchmark for world models, using Scratch projects as executable worlds. It evaluates next-state prediction, long-horizon tracking, causal attribution, and counterfactual prediction with replay-verified transitions.
- Released
- 2026-06-30
- Institutions
- University of Waterloo
- Evaluates
- Code & Software, General AI, cs.SE
- Openness
- unknown
- Importer
- Claire Radar
- Review status
- unreviewed
ScratchWorld paper
No reported scores are on record for this benchmark yet.