Patch Hierarchical Attention Transformer for Efficient Particle Jet Tagging

TL;DR

The paper introduces PHAT-JeT, a hierarchical transformer model that combines geometric message passing and patch-based attention to achieve high-accuracy jet tagging under strict computational constraints.

Contribution

It proposes a novel hierarchical transformer architecture that efficiently encodes local and global features for particle jet tagging, outperforming existing models under resource limits.

Findings

Achieves state-of-the-art accuracy on four jet tagging benchmarks.

Maintains high background rejection within limited computational budgets.

Outperforms existing resource-constrained models in efficiency and accuracy.

Abstract

Real-time jet tagging is critical for identifying short-lived particle decays in the high-throughput detectors of the Large Hadron Collider, where real-time trigger systems responsible for deciding which collision events to store impose strict latency and accuracy constraints. While transformer architectures achieve the highest jet tagging accuracy when compute is unconstrained, their quadratic self-attention cost makes inference restrictive on trigger budget. Existing efficient variants reduce the computational cost, but hinder the classification performance. To address this limitation, we introduce the Patch Hierarchical Attention Transformer (PHAT-JeT), which combines two mechanisms: a physics-inspired geometric message-passing module that encodes local detector-plane structure, and a hierarchical patch-based attention scheme that computes exact attention within small particle groups while preserving global context through lightweight patch-token communication. Within a restricted budget, PHAT-JeT achieves state-of-the-art accuracy and background rejection among all resource-constrained jet tagging models on four benchmarks (\textsc{hls4ml}, JetClass, Top Tagging, and Quark--Gluon). Our code is available at https://github.com/aaronw5/PHAT-JeT.