EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

ToR-Lite: A Lightweight Semantic Query Decomposition for Multi-Hop Retrieval-Augmented Generation in Cloud-Based AI Systems

Hee-Kyong Yoo, Wonbae Kim, Nammee Moon

Applied Sciences · 19.04.2026 · 16 (8) · 3966

DOI: 10.3390/app16083966 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 19.04.2026. Verfasst von Hee-Kyong Yoo und 2 weiteren Autor:innen. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Cloud-based AI systems increasingly rely on Retrieval-Augmented Generation (RAG) to handle complex, knowledge-intensive queries. However, query decomposition for multi-hop retrieval—traditionally powered by large language models (LLMs)—incurs significant latency and cost, rendering it impractical for large-scale, cost-sensitive cloud deployments. We propose ToR-Lite, a lightweight, generative LLM-free semantic query decomposition framework designed to enhance multi-hop retrieval efficiency in cloud-based AI systems. ToR-Lite employs a novel Word-Window Splitting algorithm that detects semantic breakpoints via sliding window embeddings, effectively decomposing complex queries without expensive LLM inference. Experiments on the MultiHop-RAG benchmark (n = 2255) demonstrate that ToR-Lite achieves +6.03 pp Hits@10 and +0.89 pp Exact Match improvements over the Baseline, while operating 3.18 times faster than LLM-based Adaptive ToR. Retrieval performance correlates monotonically with decomposition granularity: three sub-query decompositions (#Dq = 3) yields a +7.00 pp Hits@10 improvement, confirming that semantic granularity is a key driver of retrieval performance. Comparison with rule-based Baselines confirms that these gains derive from the precision of semantic boundary detection rather than decomposition quantity alone. ToR-Lite delivers nearly twice the retrieval improvement per unit of computational cost, offering a practical and cost-effective solution for latency-sensitive cloud AI deployments.

Zitieren

APA

Yoo, H., Kim, W., & Moon, N. (2026). ToR-Lite: A Lightweight Semantic Query Decomposition for Multi-Hop Retrieval-Augmented Generation in Cloud-Based AI Systems. Applied Sciences, 16(8), 3966. https://doi.org/10.3390/app16083966

Vancouver

Yoo H, Kim W, Moon N. ToR-Lite: A Lightweight Semantic Query Decomposition for Multi-Hop Retrieval-Augmented Generation in Cloud-Based AI Systems. Applied Sciences. 2026;16(8):3966. doi:10.3390/app16083966

BibTeX
@article{Yoo2026,
  author = {Yoo, Hee-Kyong and Kim, Wonbae and Moon, Nammee},
  title = {ToR-Lite: A Lightweight Semantic Query Decomposition for Multi-Hop Retrieval-Augmented Generation in Cloud-Based AI Systems},
  journal = {Applied Sciences},
  year = {2026},
  volume = {16},
  number = {8},
  pages = {3966},
  doi = {10.3390/app16083966}
}
Kostenlos starten →