energy efficiency 1fast hadamard transform 1hardware accelerator 1large language models 1low-bit quantization 1
From the 1 of 3 papers with an AI index.
6 citations
3 papers
cs.AR2026★ 6 cited
LightRot: A Light-Weighted Rotation Scheme and Architecture for Accurate Low-Bit Large Language Model Inference
Sangjin Kim, Yuseon Choi, Jungjun Oh +2
LightRot introduces a lightweight rotation scheme and a dedicated hardware accelerator that enable energy‑efficient, low‑bit inference for large language models such as LLaMA2‑13B…
cond-mat.mtrl-sci2026
Role of surface states and band modulations in ultrathin ruthenium interconnects
Gyungho Maeng, Subeen Lim, Mi Gyoung Lee +3
Mitigating the RC delay from transistor miniaturization is essential for next-generation devices, driving a focus on interconnect electrical performance. Current copper-based inter…
cond-mat.mtrl-sci2026
First-principles high-throughput screening of ruthenium compounds for advanced interconnects
Gyungho Maeng, Subeen Lim, Bonggeun Shong +1
As interconnect dimensions continue to shrink, the industry-standard copper faces a critical increase in resistivity, presenting a significant hurdle to overall device performance.…