A wet towel is twisted tight in two hands. Deep squeezing sounds mix with water dripping out onto the floor.
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 wet towel
- two hands
- a floor
Event: a wet towel is twisted between two hands
Audio presence
- a wet towel
- water droplets
Sound: ['towel squeeze', 'water dripping']
Key standards
Visual physical commonsense (V-PC)
- Two hands visibly twist a wet towel during the clip.
- Water visibly drips out of the towel and falls onto the floor as a result of the twisting.
Audio physical commonsense (A-PC)
- The dominant audible sound is a deep fabric-squeeze character alongside discrete water drip sounds.
Cross-modal physical commonsense (AV-PC)
- The squeezing sound corresponds in time to the visible twist of the towel.
- The drip sounds correspond in time to the visible water falling to the floor.