SASDenSebLE: A Compact Vision Transformer Inference Architecture with Saturation-Approximate Softmax Dataflow Enabling Sequence-Parallelism Boosted Layer-Fusion Execution
IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems(2025)
关键词
Transformer,SW-HW Co-design,Approximate Softmax,Scalability
AI 理解论文
溯源树
样例

生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要