Video will be especially important for language models to grasp physical actions that are instinctive and obvious to humans but not explicitly detailed in text

Video will be especially important for language models to grasp physical actions that are instinctive and obvious to humans but not explicitly detailed in text or video captions. I mentioned this in 2022: https://twitter.com/LechMazur/status/1607929403421462528 https://twitter.com/LechMazur/status/1607929403421462528 https://twitter.com/LechMazur/status/1619032477951213568 https://twitter.com/LechMazur/status/1619032477951213568