Fengshi Tian

About

Fengshi Tian is a rising star in the field of neuromorphic computing and energy-efficient artificial intelligence hardware, with a focus on embodied intelligence systems. His research centers on developing specialized accelerators that bridge the gap between advanced deep learning models and real-time, low-power processing for autonomous robots and interactive agents. Tian’s most notable contribution is the design of a 28nm Spiking Vision Transformer Accelerator, featuring a dual-path sparse compute core and an EMA-free self-attention engine. This work addresses critical bottlenecks in deploying vision transformers for embodied AI tasks, such as dynamic visual perception and decision-making, by dramatically reducing energy consumption while maintaining high accuracy. Although published in 2025, this pioneering paper has already garnered 2 citations, signaling its early impact in a rapidly evolving field. Tian’s achievements highlight his ability to integrate spiking neural networks with transformer architectures, offering a path toward more efficient, brain-inspired hardware for next-generation autonomous systems. His work is particularly relevant for students and researchers exploring the intersection of neuromorphic engineering, computer vision, and edge AI.

Research Focus

Key Achievements

Top Papers

1. 1
   A 28nm Spiking Vision Transformer Accelerator with Dual-Path Sparse Compute Core and EMA-free Self-Attention Engine for Embodied Intelligence

   2 citations · 2025

Key Collaborators

Contact & Links