Search papers, labs, and topics across Lattice.
4
0
9
GigaBrain-0.7 achieves unprecedented task adaptability in embodied agents, outperforming previous models in zero-shot capabilities and instruction following.
Steering personality in LLMs is not just about traits; it's a complex interplay of cognitive processes that reveals surprising geometric relationships in activation space.
MR-ConceptGCN reveals that integrating multi-relational graph structures with sequential learning can dramatically enhance user modeling in personalized systems.
LLMs struggle with adaptive planning, achieving only 67.75% accuracy when faced with progressively revealed world and user constraints.
