Max Robotics — Global Service Robot Directory + US Certification

About

Lixiang Yin is a leading researcher in computer architecture, specializing in domain-specific accelerators for machine learning and robotics. Their most prominent contribution is the introduction of NeCTAr (Near-Cache Transformer Accelerator), a groundbreaking 16nm heterogeneous multicore RISC-V system-on-chip (SoC) detailed in their 2024 paper. This design uniquely integrates both near-core and near-memory accelerators to efficiently handle sparse and dense machine learning kernels, achieving 109 GOPS for matrix-vector multiplications at just 400MHz and 0.85V. Yin’s work directly addresses the critical challenge of energy-efficient inference for large language models and real-time robotics applications. With their prototype chip demonstrating a compelling balance of performance and power efficiency, Yin has established a new paradigm for class SoCs in the AI era. Their research, which has already garnered early citations, is poised to influence next-generation edge computing and embedded AI systems, making them a rising authority in the intersection of VLSI design and intelligent systems.