Score-LM improves radar fault diagnosis with compact language interface
SCORE-LM: State-Space Radar Representations with Language Models for Fault Diagnosis
Machine Learning
Summary
Radar systems can have hardware faults that make it hard for machines to understand their surroundings. The authors created SCORE-LM, a tool that first summarizes radar data into a compact form and then uses a language model to identify faults and suggest maintenance steps. This approach is much smaller and more accurate than previous methods and can even work well with very little labelled data. The language model helps explain the faults in a way that supports clearer inspection guidance.
What this means in practice
- •For radar system engineers: Integrate compact SCORE-LM to automatically detect and classify hardware faults in radar systems for improved diagnostic accuracy and reduced computational load.
- •For maintenance teams: Use the language-guided reports from SCORE-LM to get fault-specific inspection advice that clarifies next steps for repairing radar equipment.
Authors
Mainak Mallick, Seung-Kyum Choi
Abstract
Radar hardware faults threaten automated perception, motivating accurate, compact diagnosis and understandable maintenance guidance. We introduce SCORE-LM, which couples a small scatterer-conditioned operator-response encoder (SCORE) to an adapted local language model. SCORE combines self-referenced complex trajectories, physical descriptors, and a selective state-space branch, with source-only self-supervision and directional fault inference. On eight capture-excluded Rad-R fault recordings, it achieves state-of-the-art performance within the evaluated nine-model comparison: 88.39% mean capture recall and 88.20% four-fault macro-F1 at ten frames. Its 39,520 radar inference coefficients are 119.7 times fewer than RadrNet-DS-CI's, while recall is 15.56 percentage points higher than this strongest competitor. In a separate low-label protocol, SCORE reaches 71.58% recall with one labeled source window per class. A nonlinear projector converts four frozen fault similarities into five soft tokens, linking compact diagnosis to class-conditioned maintenance guidance. On 75 development questions covering 24 radar windows, language adaptation raises correct-fault answers from 45 to 62 (60.0% to 82.7%) relative to removing the co-trained adapters, while retaining the same projector. SCORE-LM thus combines a compact radar specialist with a language interface for communicating fault-specific inspection guidance.