Search papers, labs, and topics across Lattice.
Affiliation:
3
0
4
Even when provided with complete authorization proofs in-context, LLM planners still execute unsafe mutations up to 75% of the time, proving that agent authority must be enforced deterministically at the execution boundary rather than delegated to model reasoning.
Agent-UCT slashes logical search costs by over 73% while optimizing agentic workflows, making it a game-changer for RAG pipelines.
StateFuse reveals that preserving contradictions in memory can enhance safety and correction mechanisms in multi-agent systems, challenging the conventional wisdom of collapsing conflicting information for simplicity.
