做模型量化和推理加速的团队终于有了更优的缩放策略——ScaleSearch直接提升精度且兼容现有方法，建议做低精度部署的开发者试试。
量化是加速生成模型推理的标准技术，但传统块浮点（BFP）格式使用基于块最大幅度的固定缩放因子，可能导致量化误差次优。本文提出ScaleSearch方法，通过细粒度搜索利用微缩放格式的尾数位，最小化量化误差。ScaleSearch可集成于后训练量化（PTQ）和低精度注意力机制，实验显示NVFP4量化误差降低27%，Qwen3-8B在MATH500上PTQ提升15点。此外，ScaleSearchAttention算法在Llama 3.1 70B上实现Wikitext-2困惑度降低0.77点，几乎无性能损失。
Search Your Block Floating Point Scales!
Quantization has emerged as a standard technique for accelerating inference for generative models by enabling faster low-precision computations and reduced memory transfers. Recently, GPU accelerators have added first-class support for microscaling Block Floating Point (BFP) formats. Standard BFP algorithms use a fixed scale based on the maximum magnitude of the block. We observe that this scale choice can be suboptimal with respect to quantization errors. In this work, we propose ScaleSearch, an alternative strategy for selecting these scale factors: using a fine-grained search leveraging the mantissa bits in microscaling formats to minimize the quantization error for the given distribution. ScaleSearch can be integrated with existing quantization methods such as Post Training Quantization and low-precision attention, and is shown to improve their performance. Additionally, we introduce ScaleSearchAttention, an accelerated NVFP4-based attention algorithm, which uses ScaleSearch and adapted prior techniques to ensure near-0 performance loss for causal language modeling. Experiments show that ScaleSearch reduces quantization error by 27% for NVFP4 and improves language model PTQ by up to 15 points for MATH500 (Qwen3-8B), while ScaleSearchAttention improves Wikitext-2 PPL by upto 0.77 points for Llama 3.1 70B. The proposed methods closely match baseline performance while providing quantization accuracy improvements.