Non-linear activation functions play a pivotal role in on-device inference and training, as they not only consume substantial hardware resources but also impose a significant impact on system performance and energy efficiency. In this work, we propose Distribution-Aware Piecewise Activation (DAPA), a differentiable and hardware-friendly activation function for Transformer architectures by exploiting the distribution of pre-activation data. DAPA employs a non-uniform piecewise approximation that allocates finer segments to high-probability regions of the distribution, improving generalizability over prior piecewise linear methods. The resulting approximation is further quantized using Distribution-Weighted Mean Square Error to reduce latency and resource utilization for hardware deployment. Our HLS implementation demonstrates that DAPA speeds up GELU computation by 16× and decreases DSP utilization by 16× while maintaining comparable or better performance across vision Transformers and GPT-2 models.
@article{arxiv.2603.19338,
title = {DAPA: Distribution Aware Piecewise Activation Functions for On-Device Transformer Inference and Training},
author = {Maoyang Xiang and Bo Wang},
journal= {arXiv preprint arXiv:2603.19338},
year = {2026}
}