-
ELMoE-3D: Leveraging Intrinsic Elasticity of MoE for Hybrid-Bonding-Enabled Self-Speculative Decodin…
Yuseon Choi, Jingu Lee, Jungjun Oh, Sunjoo Whang, Byeongcheol Kim, Minsung Kim, Junhyuk Lee, Hoi-Jun Yoo, Sangjin Kim
-
VectorFLASH: A 2.8-8.7 Tokens/s Resource-Efficient Long-Context LLM Accelerator with Dual Vector Qua…
Jungwan Lee, Seongyon Hong, Youngjin Moon, Jingu Lee, Junhyuk Lee, Yuseon Choi, Jungjun Oh, and Hoi-Jun Yoo
-
NeuLPU: A 3.6 mW Spiking LLM Processing Unit for On-Device Long-Context Inference
Sangmyoung Lee, Sangwoo Ha, Youngjin Moon, Minsung Kim, Seryeong Kim, Wooyoung Jo, Junhyuk Lee, Sangyeob Kim, and Hoi-Jun Yoo