Shiming Li

PhD student at Department of Information Technology; Division of Computer Systems

Telephone:: +46 18 471 10 40
E-mail:: shiming.li@it.uu.se
Visiting address:: Hus 10, Regementsvägen 10
Postal address:: Box 524
751 20 UPPSALA

Download contact information for Shiming Li at Department of Information Technology; Division of Computer Systems

Publications

Recent publications

Exploiting Unused Storage Space for Ultra-Low-Power Neural Network Inference

Li, Shiming; Åkerhielm, Sebastian; Petruccelli, Alessio; Chen, Xiaoyue et al.

Part of IEEE Embedded Systems Letters (ESL), p. 332-335, 2026
- DOI for Exploiting Unused Storage Space for Ultra-Low-Power Neural Network Inference
CONVREFLEX: Efficient Ultra-Low-Power CNN Inference via Clamping Prediction

Li, Shiming; Yao, Yuan; Mottola, Luca; Kaxiras, Stefanos

Part of SenSys '26, p. 452-464, 2026
- DOI for CONVREFLEX: Efficient Ultra-Low-Power CNN Inference via Clamping Prediction
- Download full text (pdf) of CONVREFLEX: Efficient Ultra-Low-Power CNN Inference via Clamping Prediction

All publications

Articles in journal

Exploiting Unused Storage Space for Ultra-Low-Power Neural Network Inference

Li, Shiming; Åkerhielm, Sebastian; Petruccelli, Alessio; Chen, Xiaoyue et al.

Part of IEEE Embedded Systems Letters (ESL), p. 332-335, 2026
- DOI for Exploiting Unused Storage Space for Ultra-Low-Power Neural Network Inference

Conference papers

CONVREFLEX: Efficient Ultra-Low-Power CNN Inference via Clamping Prediction

Li, Shiming; Yao, Yuan; Mottola, Luca; Kaxiras, Stefanos

Part of SenSys '26, p. 452-464, 2026
- DOI for CONVREFLEX: Efficient Ultra-Low-Power CNN Inference via Clamping Prediction
- Download full text (pdf) of CONVREFLEX: Efficient Ultra-Low-Power CNN Inference via Clamping Prediction