Search papers, labs, and topics across Lattice.
1
0
3
LLMs can now plan complex multi-robot tasks with 15% higher success rates on vague instructions by optimizing prompts with textual gradients and sharing meta-prompts across agents.
