Spatially Aware Linear Transformer (SAL-T) for Efficient Particle Jet Identification

ICLR 2026 Conference Submission21724 Authors

Keywords: Transformer, Efficient, Linear, Linformer, Physics, Particle, Machine Learning, High Energy Physics, Jet Tagging

TL;DR: SAL-T is a physics-inspired linear transformer combining kinematic feature–based spatial partitioning with convolutional token mixing to achieve near full-attention accuracy in jet classification with linear complexity and low-latency inference.

Abstract: Transformers are very effective in capturing both global and local correlations within high-energy particle collisions, but they present deployment challenges in high-data-throughput environments, such as the CERN LHC. The quadratic complexity of transformer models demands substantial resources and increases latency during inference. In order to address these issues, we introduce Spatially Aware Linear Transformer (SAL-T), a physics-inspired enhancement of the Linformer architecture that maintains linear attention. Our method incorporates spatially aware partitioning of particles based on kinematic features, thereby computing attention between regions of physical significance. Additionally, we employ convolutional layers to capture local correlations, informed by insights from jet physics. In addition to outperforming the standard Linformer in jet classification tasks, SAL-T also achieves classification results comparable to full-attention transformers, while using considerably fewer resources with lower latency during inference. Experiments on a generic point cloud classification dataset (ModelNet10) further confirm this trend.

Primary Area: applications to physical sciences (physics, chemistry, biology, etc.)

Submission Number: 21724