Teaching large language models to see in radar: aspect-distributed prototypes for few-shot HRRP ATR

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Bi, De, Xu, Chengbai, Chen, Lingfeng, Hu, Panhe
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866914423885529088
author Bi, De
Xu, Chengbai
Chen, Lingfeng
Hu, Panhe
author_facet Bi, De
Xu, Chengbai
Chen, Lingfeng
Hu, Panhe
contents High-resolution range profiles (HRRPs) play a critical role in automatic target recognition (ATR) due to their richinformationregarding target scattering centers (SCs), which encapsulate the geometric and electromagnetic characteristics of thetarget.Under few-shot circumstances, traditional learning-based methods often suffer from overfitting and struggle togeneralizeeffectively. The recently proposed HRRPLLM, which leverages the in-context learning (ICL) capabilities of largelanguagemodels (LLMs) for one-shot HRRP ATR, is limited in few-shot scenarios. This limitation arises because it primarilyutilizesthe distribution of SCs for recognition while neglecting the variance of the samples caused by aspect sensitivity. Thispaperproposes a straightforward yet effective Aspect-Distributed Prototype (ADP) strategy for LLM-based ATRunder few-shotconditions to enhance aspect robustness. Experiments conducted on both simulated and measured aircraft electromagneticdatasets demonstrate that the proposed method significantly outperforms current benchmarks.
format Preprint
id arxiv_https___arxiv_org_abs_2512_06617
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Teaching large language models to see in radar: aspect-distributed prototypes for few-shot HRRP ATR
Bi, De
Xu, Chengbai
Chen, Lingfeng
Hu, Panhe
Signal Processing
High-resolution range profiles (HRRPs) play a critical role in automatic target recognition (ATR) due to their richinformationregarding target scattering centers (SCs), which encapsulate the geometric and electromagnetic characteristics of thetarget.Under few-shot circumstances, traditional learning-based methods often suffer from overfitting and struggle togeneralizeeffectively. The recently proposed HRRPLLM, which leverages the in-context learning (ICL) capabilities of largelanguagemodels (LLMs) for one-shot HRRP ATR, is limited in few-shot scenarios. This limitation arises because it primarilyutilizesthe distribution of SCs for recognition while neglecting the variance of the samples caused by aspect sensitivity. Thispaperproposes a straightforward yet effective Aspect-Distributed Prototype (ADP) strategy for LLM-based ATRunder few-shotconditions to enhance aspect robustness. Experiments conducted on both simulated and measured aircraft electromagneticdatasets demonstrate that the proposed method significantly outperforms current benchmarks.
title Teaching large language models to see in radar: aspect-distributed prototypes for few-shot HRRP ATR
topic Signal Processing
url https://arxiv.org/abs/2512.06617