Search papers, labs, and topics across Lattice.
3
0
6
Adaptive page-aware scheduling can boost LLM serving throughput by over 1.3x on commodity GPUs, challenging existing optimization paradigms.
PCAF achieves a remarkable balance of speed and accuracy, processing over 0.61M tokens/s while outperforming dense Transformers in perplexity on long-context language modeling tasks.
LLMs in games aren't just about smarter NPCs; they fundamentally alter gameplay variability and personalization, but at the cost of predictable difficulty and coherent structure.
