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

2025/06/03 by Lingfeng Chen, Panhe Hu, Chen, Lingfeng +7
Computer Science · Medicine · #Artificial Intelligence in Healthcare and Education #FOS: Electrical engineering #Natural Language Processing Techniques #Signal Processing (eess.SP) #Topic Modeling #electronic engineering #information engineering

paper · doi:10.48550/arxiv.2506.02465

openalex publication_date 2025/06/03 · openalex created_date 2025/10/14 · openalex updated_date 2026/07/28

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.

Citations

• MaDiNet: Mamba Diffusion Network for SAR Target Detection
• HRRPGraphNet: Make HRRPs to Be Graphs for Efficient Target Recognition
• SARATR-X: Toward Building A Foundation Model for SAR Target Recognition
• SARATR-X: Toward Building a Foundation Model for SAR Target Recognition
• SARDet-100K: Towards Open-Source Benchmark and ToolKit for Large-Scale SAR Object Detection
• Time Series Forecasting with LLMs: Understanding and Enhancing Model Capabilities
• Empowering Time Series Analysis with Large Language Models: A Survey
• InfMAE: A Foundation Model in the Infrared Modality
• Large Language Models Are Zero-Shot Time Series Forecasters
• RemoteCLIP: A Vision Language Foundation Model for Remote Sensing
• Learning Transferable Visual Models From Natural Language Supervision
• What Makes Good In-Context Examples for GPT-3?
• Prototypical Networks for Few-shot Learning
• Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
• Matching Networks for One Shot Learning
• TableTime: Reformulating Time Series Classification as Training-Free Table Understanding with Large Language Models

Related

• Prompting Large Language Models for Zero-Shot Domain Adaptation in Speech Recognition
• Scaling ASR Improves Zero and Few Shot Learning
• Speech Recognition Rescoring with Large Speech-Text Foundation Models
• Effect and Analysis of Large-scale Language Model Rescoring on Competitive ASR Systems
• HRM-Text: Efficient Pretraining Beyond Scaling