Search papers, labs, and topics across Lattice.
Affiliation:
3
0
7
PATTON, a PIM runtime that integrates production LLM serving engines with commodity PIM, achieves an average 1.95x speedup and 4.83x higher energy efficiency over evaluated baselines, requires no PIM processing-unit modifications, and maintains a KV cache hit rate comparable to the native GPU KV cache in vLLM.
Adaptive imagination can transform limited target data into reliable rollouts, enabling robust sim-to-real transfer in visual reinforcement learning.
Selective layer looping in masked diffusion language models unlocks surprising gains in both training efficiency and reasoning performance, beating naive depth scaling.
