AI News

AFCE for Vision–Language–Action: predict future physical Effects, then decode to actions — how to get started

ABC2 min read
Model Releases