EUVIMEDEUVIMED entdecken
Crossref

Efficient Industrial Dataset Distillation With Textual Trajectory Matching

Muquan Li, Qian Dong, Dongyang Zhang, Ke Qin, Guangchun Luo

IEEE Transactions on Industrial Informatics · 2026-06 · 22 (6) · 5357-5368

DOI: 10.1109/tii.2026.3662409 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Industrial Informatics, 2026-06. Verfasst von Muquan Li und 4 weiteren Autor:innen. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 29 Literaturangaben. Gefördert von: Natural Science Foundation of Sichuan Province.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Modern industrial environments generate massive streams of discrete textual logs that record device status, error codes, and operational events. Deploying models on resource-constrained edge devices demands extreme data compression and fast inference, yet existing dataset distillation (DD) methods are designed for images and capture only short-term training dynamics via single-step gradient or distribution matching. To address these limitations, we propose textual trajectory matching (TTM), a novel textual DD framework that aligns student trajectories with expert trajectories derived from industrial data training. We introduce a Mask-and-Fill initialization to enhance trajectory diversity, expanding semantic representations. We further propose a manifold distribution-based initialization to preserve original semantic features through low-dimensional manifold analysis. To address model computational costs, we design a subset matching strategy to reducing by aligning critical structural components. Evaluated on SST-2, MNLI-m, AGNews, and BlueGene/L (BGL) with $\text{BERT}_{\text{BASE}}$ , $\text{RoBERTa}_{\text{BASE}}$ , $\text{XLNet}_{\text{BASE}}$ , and LLaMA 3, TTM achieves 3.6% higher accuracy than state-of-the-art methods, enabling efficient industrial dataset and model deployment in smart factories.

Förderung

  • Natural Science Foundation of Sichuan Province

Zitieren

APA

Li, M., Dong, Q., Zhang, D., Qin, K., & Luo, G. (2026). Efficient Industrial Dataset Distillation With Textual Trajectory Matching. IEEE Transactions on Industrial Informatics, 22(6), 5357-5368. https://doi.org/10.1109/tii.2026.3662409

Vancouver

Li M, Dong Q, Zhang D, Qin K, Luo G. Efficient Industrial Dataset Distillation With Textual Trajectory Matching. IEEE Transactions on Industrial Informatics. 2026;22(6):5357-5368. doi:10.1109/tii.2026.3662409

BibTeX
@article{Li2026,
  author = {Li, Muquan and Dong, Qian and Zhang, Dongyang and Qin, Ke and Luo, Guangchun},
  title = {Efficient Industrial Dataset Distillation With Textual Trajectory Matching},
  journal = {IEEE Transactions on Industrial Informatics},
  year = {2026},
  volume = {22},
  number = {6},
  pages = {5357--5368},
  doi = {10.1109/tii.2026.3662409}
}
Kostenlos starten →