Attn-QAT: 4-Bit Attention With Quantization-Aware Training
arXiv:2603.00040v1 Announce Type: new Abstract: Achieving reliable 4-bit attention is a prerequisite for end-to-end FP4 computation on emerging FP4-capable GPUs, yet attention remains the main …
Peiyuan Zhang, Matthew Noto, Wenxuan Tan, Chengquan Jiang, Will Lin, Wei Zhou, Hao Zhang
조회수 49회