A hand peels tape off a glass surface slowly, producing a high smooth peeling sound. Then peels it faster, producing a louder more chaotic crackle.
Seedance 2.0ByteDance · proprietary
Kling 3.0 OmniKuaishou · proprietary
Veo 3.1Google · proprietary
LTX-2.3Lightricks · open
Ovi 1.1character.ai · open
JavisDiT++JavisDiT · open
MagiHumanGAIR-NLP · open
Per-prompt rubric
Basic standards
Visual presence
- a hand
- tape
- a glass surface
Event: a hand peels tape off a glass surface
Audio presence
- tape
Sound: tape peeling from glass
Key standards
Visual physical commonsense (V-PC)
- A hand visibly peels tape off a glass surface with continuous motion showing adhesive release.
Audio physical commonsense (A-PC)
- The faster peel phase is audibly louder and more chaotic than the slower peel phase.
Cross-modal physical commonsense (AV-PC)
- The peeling sound aligns in time with the visible separation of tape from the glass surface.
- The louder, more chaotic peeling sound corresponds to the faster hand motion.
- The quieter, smoother peeling sound corresponds to the slower hand motion.