Group

Deep Efficiency Lab, School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore.

The Deep Efficiency Lab at the School of Electrical and Electronic Engineering, NTU Singapore, works on AI efficiency across the full stack, from models and algorithms down to silicon. Interested in joining? See Join Us.

Core Team

  • Dr. Ali Azarpeyvand, Research Fellow (since 09/26): digital IC design and computer architecture
  • Yanjing Li, Research Fellow (since 09/26): AI acceleration and AI infrastructure
  • Yanlong Chen, PhD student (since 08/26): model acceleration and quantization
  • Amitoj Singh, Research Assistant (since 09/26): AI accelerator design and digital IC design
  • Yifei He, Research Assistant (since 11/26): AI accelerator design and digital IC design

MSc Dissertation Students, NTU

  • 09/26-now: Xu Yifeng. Dataflow extensions of FlatAttention to state-space kernels for on-robot perception
  • 09/26-now: Meng Yishan. Agentic electronic design automation loop for small accelerator block refinement
  • 09/26-now: Liu Qingren. Near-memory processing accelerator for KV-cache-dominated attention
  • 09/26-now: Zhang Haorui. RISC-V vector extension design for low-bit embodied-AI inference
  • 09/26-now: Deng Qiuyi. RISC-V shared-L1 tile cluster for real-time robot control
  • 09/26-now: He Zihanshu. Trace-driven architecture modelling for embodied-AI accelerators
  • 09/26-now: Zhao Yikai. Dual-precision processing element for adaptation-compatible AI compute
  • 09/26-now: Chai Jiayang. Rate-distortion bit allocation for the key-value cache of vision-language-action models
  • 09/26-now: Jiang Haojun. Salience-driven mixed-precision quantization for multimodal language models
  • 09/26-now: Cao Benbo. Reconfigurable sparse matrix-multiply engine for adaptable VLA inference

Undergraduate Researchers (URECA), NTU

  • 09/26-now: Guo Yichen. Efficient multimodal world models with hardware-aware design and acceleration
  • 09/26-now: Ethan Kok Song Jie. Accelerating vision-language-action model inference on edge GPUs
  • 09/26-now: Khanuja Jaiveer Singh. Automated PPA regression study of small AI-accelerator IP blocks
  • 09/26-now: Thongthornpatch Chamadol. Post-training quantization for vision-language-action models
  • 09/26-now: Pham Ngoc Quang. Full-platform simulation of a small AI kernel on GVSoC
  • 09/26-now: Yeap Wei Han. Low-bit GEMM kernels for a RISC-V vector core

Co-supervised PhD Students

  • 10/25-06/26: Aofeng Shen, PhD student @ ETH Zürich, with Luca Benini
  • 05/25-06/26: Anna Tegon, PhD student @ ETH Zürich, with Luca Benini
  • 04/25-12/25: Enrico Vezzali, visiting PhD student @ ETH Zürich, PhD student @ University of Modena and Reggio Emilia, with Prof. Costantino Grana, Prof. Luca Benini, and Prof. Federico Bolelli
  • 10/24-06/26: Mahdi Abdollahpour, PhD student @ University of Bologna, with Alessandro Vanelli Coralli and Luca Benini
  • 10/24-06/26: Chi Zhang, PhD student @ ETH Zürich, with Luca Benini
  • 09/24-06/26: Giusy Spacone, PhD student @ ETH Zürich, with Luca Benini
  • 03/24-06/25: Morin Lucas, PhD student @ ETH Zürich, with Luc Van Gool. Now Applied Scientist @ Onfido/Entrust
  • 12/23-06/25: Thorir Mar Ingolfsson, PhD student @ ETH Zürich, with Luca Benini. Now postdoctoral researcher @ ETH Zürich
  • 08/23-02/26: Bin Ren, visiting PhD student @ ETH Zürich, with Luc Van Gool, Nicu Sebe, and Rita Cucchiara. Now postdoctoral researcher @ MBZUAI
  • 04/23-06/26: Pietro Bonazzi, PhD student @ ETH Zürich, with Michele Magno

Alumni: Master and Undergraduate Students