Search papers, labs, and topics across Lattice.
Affiliation:
3
4
6
7
Calibrating procedural relations in skill graphs boosts LLM task performance by over 10 points while streamlining retrieval efficiency.
System Intelligence emerges as a game-changer, enabling LLM agents to collaborate effectively across complex tasks by organizing their interactions through dynamic graph structures.
Tired of DPO's length bias and probability degradation? LMPO offers a more robust and efficient alternative for preference-based RLHF, outperforming existing methods on Mistral and LLaMA3.
