PARALLEL SESSION
B4L-E
Efficient Arithmetic for ML and CIM
WEDNESDAY - 10 September | 13:20 - 15:00
CHAIRS
Kavya Sreedhar, Francesco Conti
SESSION PROGRAM
13:20 - 13:45
PADCIM: A 1966.9-TOPS/W 2.09%-Rmse Probabilistic Approximate Dynamic-Logic Based Digital Computing-in-Memory Macro in 40nm
Wente Yi, Zhenyu Xue, Tianshuo Bai, Lehao Tan, Han Zhang, Jingcheng Gu, Jintao Mo, Wenjia Wang, Biao Pan, Weisheng Zh...
13:45 - 14:10
SAFER: Sparsity Integrated Compute-in-Memory AI Accelerator with a Fused Dot-Product Engine and a RISC-V CPU
Amitesh Sridharan, Asmer Hamid Ali, Yongjae Lee, Anupreetham Anupreetham, Yaotian Liu, Jeff Zhang, Jae-Sun Seo, Delia...
14:10 - 14:35
A 28nm 31.2 TOPS/W Diffusion Transformer Accelerator Using Memory-in-Compute Architecture
Soonha Hwang, Kaining Zhou, Nathan Chiang, Vignesh Sundaresha, Naresh Shanbhag
14:35 - 15:00
A 9-b 35 TOPS/W Differential Ring-Oscillator Based Multiply-and-Accumulate Macro with On-Chip PVT Calibration
Chawin Khongprasongsiri, Sayan Kumar, Fergal Kilgannon, Derek Shaw, Panassaya Ounsawatdipong, Teerachot Siriburanon, ...