Preprint
Sep 2026
Quantifying the Effect of HCLs on a Fixed-Microarchitecture MXFP4 Accelerator
This paper compares the most widely used HCLs using the same fixed design, the OCP MXFP4 block dot product, a quantization primitive at the heart of edge Physical-AI inference, implemented as a single 12-stage, II=1 pipeline.
D. Passaretti, Sajjad Tamimi, Nicola Dall'Ora
· 0 citations