EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

ET: An Energy Efficient Edge Transformer Architecture

Shashank Nag, Alan Bacellar, Anshul Jha, Eugene John, Krishnan Kailas, Priscila Lima, Neeraja Yadwadkar, Felipe França, Lizy K. John

ACM Transactions on Architecture and Code Optimization · 30.09.2026 · 23 (3) · 1-26

DOI: 10.1145/3828726 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACM Transactions on Architecture and Code Optimization, online erschienen am 24.07.2026. Verfasst von Shashank Nag und 8 weiteren Autor:innen. Die Arbeit nennt 58 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Transformers are increasingly being deployed at the edge to support agentic AI, real-time processing, and privacy-preserving applications. However, their substantial computational and memory requirements often exceed the energy and latency constraints of edge devices. Through detailed characterization, we observe that the majority of computations and parameters in transformer models are concentrated in the feed-forward Multi-Layer Perceptron (MLP) blocks, identifying them as a key target for optimization. In this work, we introduce ET, a novel transformer architecture tailored for energy-efficient edge inference. By integrating shared Look Up Table (LUT) based feed-forward blocks with traditional self-attention modules, ET significantly reduces computational load, memory footprint, and energy usage, aligning closely with the operational and resource constraints of edge devices. We present the design, implementation, and evaluation of ET, and demonstrate that it delivers comparable accuracy to traditional transformers while achieving 2.1× improvement over a baseline ASIC accelerator. It also achieves a 1.12× efficiency on GPU acceleration, and about 1.34× improvement over a prior BinaryBERT FPGA accelerator. We further report 1.2× energy efficiency over recent multiplication-free models, such as Bitnet, showing that ET offers a compelling and practical solution for deploying transformers effectively at the edge, whether on FPGAs, ASICs, or edge GPUs. 1

Zitieren

APA

Nag, S., Bacellar, A., Jha, A., John, E., Kailas, K., Lima, P., Yadwadkar, N., França, F., & John, L. K. (2026). ET: An Energy Efficient Edge Transformer Architecture. ACM Transactions on Architecture and Code Optimization, 23(3), 1-26. https://doi.org/10.1145/3828726

Vancouver

Nag S, Bacellar A, Jha A, John E, Kailas K, Lima P, et al. ET: An Energy Efficient Edge Transformer Architecture. ACM Transactions on Architecture and Code Optimization. 2026;23(3):1-26. doi:10.1145/3828726

BibTeX
@article{Nag2026,
  author = {Nag, Shashank and Bacellar, Alan and Jha, Anshul and John, Eugene and Kailas, Krishnan and Lima, Priscila and Yadwadkar, Neeraja and França, Felipe and John, Lizy K.},
  title = {ET: An Energy Efficient Edge Transformer Architecture},
  journal = {ACM Transactions on Architecture and Code Optimization},
  year = {2026},
  volume = {23},
  number = {3},
  pages = {1--26},
  doi = {10.1145/3828726}
}
Kostenlos starten →