Search papers, labs, and topics across Lattice.
3
0
5
Summary tokens are not just compressed context, but also dynamic KV cache indices, unlocking flexible memory compression and reduced compute in hybrid attention.
Non-monotonic performance in experimental trajectories reveals that later rounds can sometimes undermine earlier successes, challenging conventional assumptions in industrial ML research.
Surprisingly, the "think before answer" paradigm fails to enhance generative recommendation models, prompting a novel approach that redefines how reasoning is integrated into these systems.
