Pinned post
π€π€ Give a video generation model a scene and a goal. Can it generate egocentric execution to get the job done?
We tested 110 real-world scenarios on 6 SoTA models to see how they perform.
Introducing Ego2Act, a goal-directed video generation benchmark π
π ego2act.github.io
π§ͺ alphaxiv.org/abs/2610.01092
π€ huggingface.co/datasets/ego2aβ¦
[π§΅Thread]
We tested 110 real-world scenarios on 6 SoTA models to see how they perform.
Introducing Ego2Act, a goal-directed video generation benchmark π
π ego2act.github.io
π§ͺ alphaxiv.org/abs/2610.01092
π€ huggingface.co/datasets/ego2aβ¦
[π§΅Thread]
0:16