EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

LLM-Driven News Recommendation via Lightweight Task-Adaptive Modules

Han Wei, Tong Niu, Sisi Peng, Minchen Xu, Dan Qu

Applied Sciences · 07.07.2026 · 16 (13) · 6818

DOI: 10.3390/app16136818 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 07.07.2026. Verfasst von Han Wei und 4 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Henan Province Major Industrial “Challenge-Based Innovation”.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Large language models produce semantically rich embeddings, yet direct employment of generic LLM embeddings fails to satisfy news recommendation demands due to inherent semantic mismatches with task targets. Existing fine-tuning methods including LoRA can narrow such gaps but bring prohibitive computational overhead, restricting real-world deployment. This work proposes lightweight task-adaptive modules (TAMs). It keeps LLM parameters fixed and transforms offline embeddings into task-specialized representations without full-model backpropagation, drastically cutting training costs. Evaluated on MIND benchmarks across ten mainstream recommendation architectures, TAMs achieves comparable accuracy to LoRA, with computational cost reduced to 1/10 of LoRA’s level. TAMs outperform GloVe-based models by 1.3–12.4% in AUC and scale effectively to MINDlarge. Ablation experiments confirm that the nonlinear projection is pivotal to performance improvement, and statistical validation across three random seeds confirms result robustness. This paradigm provides an efficient low-cost solution for LLM-based news recommendation under resource constraints.

Förderung

  • Henan Province Major Industrial “Challenge-Based Innovation”

Zitieren

APA

Wei, H., Niu, T., Peng, S., Xu, M., & Qu, D. (2026). LLM-Driven News Recommendation via Lightweight Task-Adaptive Modules. Applied Sciences, 16(13), 6818. https://doi.org/10.3390/app16136818

Vancouver

Wei H, Niu T, Peng S, Xu M, Qu D. LLM-Driven News Recommendation via Lightweight Task-Adaptive Modules. Applied Sciences. 2026;16(13):6818. doi:10.3390/app16136818

BibTeX
@article{Wei2026,
  author = {Wei, Han and Niu, Tong and Peng, Sisi and Xu, Minchen and Qu, Dan},
  title = {LLM-Driven News Recommendation via Lightweight Task-Adaptive Modules},
  journal = {Applied Sciences},
  year = {2026},
  volume = {16},
  number = {13},
  pages = {6818},
  doi = {10.3390/app16136818}
}
Kostenlos starten →