# A transformer on FPGA for low latency applications in particle physics and beyond

- Institution: CERN
- Speaker: Elias Leutgeb
- Year: 2026
- Language: English
- Subjects: Physics
- Source: [CERN](https://videos.cern.ch/record/3021008)
- Licence: [CERN](https://copyright.web.cern)
- Who can watch: Tweens profiles and older (sources give no ratings; placed by Invisible Network from the source's own words)

To identify physics phenomena that are difficult to detect by standard triggers, real-time modern AI triggers are needed. One of the most promising novel trigger classes is anomaly detection triggers, capable of flagging rare events without relying on a full theoretical understanding of the underlying process. By leveraging modern, heterogeneous accelerator platforms, like the AMD Versal Adaptive Compute Acceleration Platform, which combines classic Field Programmable Gate Array (FPGA) logic with higher-level AI Engines, we aim to achieve ultra-low-latency inference at a high throughput, necessary for operations in the CMS trigger and scouting system. We present a machine learning architecture based on a Transformer model, specifically adapted for low-latency inference. As the Transformer architecture has shown great success in interpreting complex relationships across many areas of research, we believe that a fast, flexible, high-throughput Transformer model, implemented 'on the edge', has wider-reaching applications across technology and industry.

0:00:00 Slide 1

0:00:03 Slide 2

0:00:19 Slide 3

0:00:48 Slide 4

0:01:11 Slide 5

0:01:28 Slide 6

0:02:46 Slide 7

0:03:10 Slide 8

0:03:30 Slide 9

0:03:51 Slide 10

0:04:18 Slide 11

0:04:42 Slide 12

0:05:27 Slide 13

0:06:31 Slide 14

0:07:02 Slide 15

0:07:26 Slide 16

0:07:48 Slide 17

0:08:05 Slide 18

0:09:04 Slide 19

0:09:41 Slide 20

0:10:09 Slide 21

0:10:30 Slide 22

0:12:02 Slide 23

0:12:48 Slide 24

0:13:58 Slide 25

0:14:24 Slide 26

0:17:20 Slide 27r>

## Lectures

1. [A transformer on FPGA for low latency applications in particle physics and beyond](https://invisiblenetwork.ai/learn/cern/3021008/28f5cda387d2), 18 min (CERN's player)