Large Language Models Can Achieve Explainable and Training-Free One-shot HRRP ATR

TL;DR

This paper presents a novel, training-free approach for HRRP automatic target recognition using large language models, converting signals into text and leveraging pre-trained knowledge without fine-tuning.

Contribution

It introduces a pioneering framework that uses LLMs for HRRP ATR by converting signals into text and employing few-shot learning, eliminating the need for training or fine-tuning.

Findings

Achieves explainable HRRP ATR without training.

Utilizes large pre-trained language models effectively.

Codes are publicly available for further research.

Abstract

This letter introduces a pioneering, training-free and explainable framework for High-Resolution Range Profile (HRRP) automatic target recognition (ATR) utilizing large-scale pre-trained Large Language Models (LLMs). Diverging from conventional methods requiring extensive task-specific training or fine-tuning, our approach converts one-dimensional HRRP signals into textual scattering center representations. Prompts are designed to align LLMs' semantic space for ATR via few-shot in-context learning, effectively leveraging its vast pre-existing knowledge without any parameter update. We make our codes publicly available to foster research into LLMs for HRRP ATR.