Search papers, labs, and topics across Lattice.
2
0
5
4
WIZARD achieves up to 14x better performance on unseen tasks by generating task-specific adaptation weights in a single forward pass, eliminating the need for costly fine-tuning.
LLMs exhibit a surprising self-preservation bias, prioritizing their own survival even when it contradicts objective utility and security, suggesting current safety training may be insufficient.
