Showing cs.ARShow all
2 papers · 1 filter
cs.AR2026
Hardware Acceleration of Block-Diffusion LLM for Edge Devices
Wei-Hsing Huang, Kiseok Lee, Ming-Yen Lee +7
Single-stream (batch-one) edge inference cannot amortize weight traffic across requests. Full-attention diffusion LLMs recompute the entire sequence at every step; native block dif…
cs.AR2026
System-Technology Co-Optimization of Bitline Routing and Bonding Pathways in Monolithic 3D DRAM Architectures
Kiseok Lee, Sungwon Cho, Seongkwang Lim +2
3D DRAM has emerged as a promising approach for continued density scaling, but its viability is limited by routing and hybrid bonding constraints to periphery, which may degrade se…