Search papers, labs, and topics across Lattice.
1
0
2
5
Current vision-language models fail to achieve embodied self-awareness, with none surpassing a 16.8% success rate in real-world interaction tasks.
