Search papers, labs, and topics across Lattice.
1
0
3
Only one out of 37 open-weight AI models meets the necessary evaluation standards to ensure safety and robustness, revealing a critical oversight in current practices.
