SCIM: Self-Correcting Iterative Mechanism for Retrieval-Augmented Generation
Electronics · 27.02.2026 · 15 (5) · 996
DOI: 10.3390/electronics15050996 ↗Kurz gefasst
Fachartikel in der Zeitschrift Electronics, online erschienen am 27.02.2026. Verfasst von Ke Li und 1 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Social Science Foundation of China.
Abstract (Originalsprache)
Standard Retrieval-Augmented Generation (RAG) models are limited by their “one-shot” nature, failing to assess or improve answer quality dynamically. To address this, we introduce SCIM (Self-Correcting Iterative Mechanism), a framework featuring multi-dimensional evaluation and adaptive retrieval. A key distinction of SCIM is its efficiency: it operates on a lightweight Flan-T5-base model (250M parameters) and requires no fine-tuning, challenging the industry’s reliance on 7B+ parameter models. Experimental results across four major benchmarks show that SCIM yields a 17.2% improvement over standard RAG (p
Förderung
- National Social Science Foundation of China
Zitieren
APA
Li, K., & Zhang, T. (2026). SCIM: Self-Correcting Iterative Mechanism for Retrieval-Augmented Generation. Electronics, 15(5), 996. https://doi.org/10.3390/electronics15050996
Vancouver
Li K, Zhang T. SCIM: Self-Correcting Iterative Mechanism for Retrieval-Augmented Generation. Electronics. 2026;15(5):996. doi:10.3390/electronics15050996
BibTeX
@article{Li2026,
author = {Li, Ke and Zhang, Tingting},
title = {SCIM: Self-Correcting Iterative Mechanism for Retrieval-Augmented Generation},
journal = {Electronics},
year = {2026},
volume = {15},
number = {5},
pages = {996},
doi = {10.3390/electronics15050996}
}
EUVIMED