# Sub-microsecond Transformers for Jet Tagging on FPGAs

arXiv: https://arxiv.org/abs/2510.24784 · HTML: https://mlsanity.com/paper/2510.24784/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 16 of its 35 references that are in the map.
- First posted on arXiv 26 Oct 2025.

## Builds on (most cited first, 16 of 16)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Linformer: Self-Attention with Linear Complexity](https://mlsanity.com/paper/2006.04768/) (2020)
- [Particle Transformer for Jet Tagging](https://mlsanity.com/paper/2202.03772/) (2022)
- [Automatic heterogeneous quantization of deep neural networks for low-latency inference on the edge for particle detectors](https://mlsanity.com/paper/2006.10159/) (2020)
- [AutoQ: Automated Kernel-Wise Neural Network Quantization](https://mlsanity.com/paper/1902.05690/) (2019)
- [hls4ml: An Open-Source Codesign Workflow to Empower Scientific Low-Power Machine Learning Devices](https://mlsanity.com/paper/2103.05579/) (2021)
- [Compressing deep neural networks on FPGAs to binary and ternary precision with HLS4ML](https://mlsanity.com/paper/2003.06308/) (2020)
- [Lorentz-Equivariant Geometric Algebra Transformers for High-Energy Physics](https://mlsanity.com/paper/2405.14806/) (2024)
- [Explainable Equivariant Neural Networks for Particle Physics: PELICAN](https://mlsanity.com/paper/2307.16506/) (2023)
- [LL-GNN: Low Latency Graph Neural Networks on FPGAs for High Energy Physics](https://mlsanity.com/paper/2209.14065/) (2022)
- [Ultrafast jet classification on FPGAs for the HL-LHC](https://mlsanity.com/paper/2402.01876/) (2024)
- [Fast Jet Tagging with MLP-Mixers on FPGAs](https://mlsanity.com/paper/2503.03103/) (2025)
- [da4ml: Distributed Arithmetic for Real-time Neural Networks on FPGAs](https://mlsanity.com/paper/2507.04535/) (2025)
- [Ultra Fast Transformers on FPGAs for Particle Physics Experiments](https://mlsanity.com/paper/2402.01047/) (2024)
- [Fast Neural Network Inference on FPGAs for Triggering on Long-Lived Particles at Colliders](https://mlsanity.com/paper/2307.05152/) (2023)
- [Neural Architecture Codesign for Fast Physics Applications](https://mlsanity.com/paper/2501.05515/) (2025)

## Built on by (most cited first, 6 of 6)

- [hls4ml: A Flexible, Open-Source Platform for Deep Learning Acceleration on Reconfigurable Hardware](https://mlsanity.com/paper/2512.01463/) (2025)
- [JetFormer: A Scalable and Efficient Transformer for Jet Tagging from Offline Analysis to FPGA Triggers](https://mlsanity.com/paper/2601.17215/) (2026)
- [Towards Tensor Network Models for Low-Latency Jet Tagging on FPGAs](https://mlsanity.com/paper/2601.10801/) (2026)
- [HGQ-LUT: Fast LUT-Aware Training and Efficient Architectures for DNN Inference](https://mlsanity.com/paper/2604.22293/) (2026)
- [Reconfigurable Computing Challenge: Transformer for Jet Tagging on Versal AI Engines](https://mlsanity.com/paper/2606.17500/) (2026)
- [Machine learning for the LHC physics program: a 2025-2026 stocktake](https://mlsanity.com/paper/2609.32874/) (2026)

## Abstract (arXiv)

We present the first sub-microsecond transformer implementation on an FPGA achieving competitive performance for state-of-the-art high-energy physics benchmarks. Transformers have shown exceptional performance on multiple tasks in modern machine learning applications, including jet tagging at the CERN Large Hadron Collider (LHC). However, their computational complexity prohibits use in real-time applications, such as the hardware trigger system of the collider experiments up until now. In this work, we demonstrate the first application of transformers for jet tagging on FPGAs, achieving O(100) nanosecond latency with superior performance compared to alternative baseline models. We leverage high-granularity quantization and distributed arithmetic optimization to fit the entire transformer model on a single FPGA, achieving the required throughput and latency. Furthermore, we add multi-head attention and linear attention support to hls4ml, making our work accessible to the broader fast machine learning community. This work advances the next-generation trigger systems for the High Luminosity LHC, enabling the use of transformers for real-time applications in high-energy physics and beyond.