EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Evidence-Grounded Multi-Label Learning with a Large Language Model for Compound Defect Detection in Rotating Components

Xiangfu Zou, Mengjie Wang, Minghua Cui, Han Zheng, Jian Chen, Zisheng Wang

Mathematics · 18.11.2026 · 14 (22) · 4237

DOI: 10.3390/math14224237 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Mathematics, online erschienen am 18.11.2026. Verfasst von Xiangfu Zou und 5 weiteren Autor:innen. Die Arbeit nennt 46 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Concurrent defects in rotating components produce overlapping vibration components, complicating the recognition of defect combinations absent from training. We formulate this task as compositional multi-label learning and propose a compositional evidence-grounded framework (CEG-LLM). The model uses a frozen large language model as a sequence encoder, with trainable low-rank adapters and four sigmoid defect outputs. Continuous waveform tokens and nominal characteristic-order descriptors represent the signal, while a four-node composition graph conditions label-wise attention. Spectral alignment and compositional regularization guide training, and a fixed template links predictions to structured signal records. On one dataset collected from a rotating component at 1500 rpm, CEG-LLM achieves an observed-state macro-F1 of 0.941. Corresponding values are 0.922 for a one-dimensional vision Transformer (ViT), 0.912 for a signal-to-text language encoder, and 0.900 for a convolutional neural network (CNN). Mean exact match across six withheld defect pairs is 0.767, and macro-F1 at −3 dB is 0.819. A top-token versus random-mask deletion contrast of 0.207 indicates sensitivity to the selected representation under that intervention. These results assess recognition of unseen defect combinations and representation sensitivity within the evaluated operating condition.

Zitieren

APA

Zou, X., Wang, M., Cui, M., Zheng, H., Chen, J., & Wang, Z. (2026). Evidence-Grounded Multi-Label Learning with a Large Language Model for Compound Defect Detection in Rotating Components. Mathematics, 14(22), 4237. https://doi.org/10.3390/math14224237

Vancouver

Zou X, Wang M, Cui M, Zheng H, Chen J, Wang Z. Evidence-Grounded Multi-Label Learning with a Large Language Model for Compound Defect Detection in Rotating Components. Mathematics. 2026;14(22):4237. doi:10.3390/math14224237

BibTeX
@article{Zou2026,
  author = {Zou, Xiangfu and Wang, Mengjie and Cui, Minghua and Zheng, Han and Chen, Jian and Wang, Zisheng},
  title = {Evidence-Grounded Multi-Label Learning with a Large Language Model for Compound Defect Detection in Rotating Components},
  journal = {Mathematics},
  year = {2026},
  volume = {14},
  number = {22},
  pages = {4237},
  doi = {10.3390/math14224237}
}
Kostenlos starten →