一句话：提出值级并行计算范式与 Carat 架构，通过时序编码复用唯一乘积，将乘法转为累加，实现无乘法器 GEMM。
针对深度神经网络中通用矩阵乘法（GEMM）的硬件加速，随着批处理和低精度数据（如 FP8）的普及，值复用潜力尚未被充分挖掘。本文提出一种新的计算范式——值级并行（value-level parallelism），即唯一乘积只计算一次，不同输入通过时序编码（temporal coding）订阅（选择）所需乘积。基于此设计的架构 Carat 将乘法转化为累加，以高效的无乘法器硬件执行 GEMM。实验表明，在等面积下，Carat 相比脉动阵列平均提升吞吐量 1.02 倍、能效 1.06 倍；扩展到多节点时，吞吐量和能效分别提升 3.2 倍和 4.3 倍。
In recent years, hardware architectures optimized for general matrix multiplication (GEMM) have been well studied to deliver better performance and efficiency for deep neural networks. With trends towards batched, low-precision data, e.g., FP8 format in this work, we observe that there is growing untapped potential for value reuse. We propose a novel computing paradigm, value-level parallelism, whereby unique products are computed only once, and different inputs subscribe to (select) their products via temporal coding. Our architecture, Carat, employs value-level parallelism and transforms multiplication into accumulation, performing GEMMs with efficient multiplier-free hardware. Experiments show that, on average, Carat improves iso-area throughput and energy efficiency by 1.02× and 1.06× over a systolic array and 3.2× and 4.3× when scaled up to multiple nodes.