@article{ART003258695},
author={Hee-Kyong Yoo and Nammee Moon},
title={ToR-RAG: A Tree-of-Retrieval-based Retrieval-Augmented Generation for Complex Question Processing},
journal={Journal of The Korea Society of Computer and Information},
issn={1598-849X},
year={2025},
volume={30},
number={10},
pages={23-31}
TY - JOUR
AU - Hee-Kyong Yoo
AU - Nammee Moon
TI - ToR-RAG: A Tree-of-Retrieval-based Retrieval-Augmented Generation for Complex Question Processing
JO - Journal of The Korea Society of Computer and Information
PY - 2025
VL - 30
IS - 10
PB - The Korean Society Of Computer And Information
SP - 23
EP - 31
SN - 1598-849X
AB - Large Language Models have advanced natural language understanding, yet remain limited in handling complex multi-hop queries requiring integration across multiple documents. Traditional Retrieval-Augmented Generation adopts a linear query-retrieval-generation pipeline, which often causes error propagation, incomplete evidence coverage, and reduced reliability. To overcome these issues, this study proposes a Tree-of-Retrieval based RAG (ToR-RAG). ToR-RAG decomposes queries into binary sub-queries via LLM prompting, performs retrieval and partial answer generation at each branch, and evaluates outputs using an LLM-as-a-Judge module. Branches below a quality threshold are pruned, ensuring efficiency. An MMR-based retrieval strategy (λ=0.75) with top-k=5 selection balances relevance and diversity. Experiments on the MultiHop-RAG dataset show that ToR-RAG improves Exact Match by +6.42 and F1 by +6.04 compared to Non-RAG, Native-RAG, and CoR-RAG. Performance peaked at Depth=3, while Depth=4 caused degradation from excessive branching and token usage. These results demonstrate that ToR-RAG enhances both accuracy and reliability in multi-hop reasoning, suggesting applicability in domains such as policy analysis, medical decision-making, and financial risk assessment.
KW - RAG;Tree of Retrieval;Multi-Hop QA;Query Decomposition;Structured Reasoning
DO -
UR -
ER -
Hee-Kyong Yoo and Nammee Moon. (2025). ToR-RAG: A Tree-of-Retrieval-based Retrieval-Augmented Generation for Complex Question Processing. Journal of The Korea Society of Computer and Information, 30(10), 23-31.
Hee-Kyong Yoo and Nammee Moon. 2025, "ToR-RAG: A Tree-of-Retrieval-based Retrieval-Augmented Generation for Complex Question Processing", Journal of The Korea Society of Computer and Information, vol.30, no.10 pp.23-31.
Hee-Kyong Yoo, Nammee Moon "ToR-RAG: A Tree-of-Retrieval-based Retrieval-Augmented Generation for Complex Question Processing" Journal of The Korea Society of Computer and Information 30.10 pp.23-31 (2025) : 23.
Hee-Kyong Yoo, Nammee Moon. ToR-RAG: A Tree-of-Retrieval-based Retrieval-Augmented Generation for Complex Question Processing. 2025; 30(10), 23-31.
Hee-Kyong Yoo and Nammee Moon. "ToR-RAG: A Tree-of-Retrieval-based Retrieval-Augmented Generation for Complex Question Processing" Journal of The Korea Society of Computer and Information 30, no.10 (2025) : 23-31.
Hee-Kyong Yoo; Nammee Moon. ToR-RAG: A Tree-of-Retrieval-based Retrieval-Augmented Generation for Complex Question Processing. Journal of The Korea Society of Computer and Information, 30(10), 23-31.
Hee-Kyong Yoo; Nammee Moon. ToR-RAG: A Tree-of-Retrieval-based Retrieval-Augmented Generation for Complex Question Processing. Journal of The Korea Society of Computer and Information. 2025; 30(10) 23-31.
Hee-Kyong Yoo, Nammee Moon. ToR-RAG: A Tree-of-Retrieval-based Retrieval-Augmented Generation for Complex Question Processing. 2025; 30(10), 23-31.
Hee-Kyong Yoo and Nammee Moon. "ToR-RAG: A Tree-of-Retrieval-based Retrieval-Augmented Generation for Complex Question Processing" Journal of The Korea Society of Computer and Information 30, no.10 (2025) : 23-31.
