Search papers, labs, and topics across Lattice.
3
0
6
8
Cheaper LLM judges can match the performance of their more expensive counterparts in citation verification, challenging the assumption that only high-cost models are suitable for deep-research tasks.
Harness recursion boosts coding agent performance by over 9% without changing the underlying model, showcasing a powerful new approach to long-context reasoning.
LLMs can now remember and reason about long-term conversations with significantly improved accuracy thanks to a new temporal-aware memory framework that structures dialogue into event calendars.
