Search papers, labs, and topics across Lattice.
1
0
3
Frozen LLMs can autonomously achieve recursive self-improvement by learning and refining explicit, executable world models from interaction data.
