EUVIMEDEUVIMED entdecken
Open Access · CC BY-SA 4.0

Improving Tree Traversal in Hierarchical RAG via Reasoning-Aware Approach with Interleaving Chain-of-Thought

Mukhamad Khafid Maassobirin, Wahyu Syaifullah J.S., Andri Fauzan Adziima

bit-Tech · 10.08.2026 · 9 (1) · 30-40

DOI: 10.32877/bt.v9i1.3785 ↗

Kurz gefasst

Fachartikel in der Zeitschrift bit-Tech, online erschienen am 10.08.2026. Verfasst von Mukhamad Khafid Maassobirin und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Retrieval-Augmented Generation (RAG) has improved the factual grounding of large language models; however, conventional retrieval strategies remain limited for long-document question answering, as relevant information is often distributed across multiple document sections and may be inferential rather than lexically similar to the query. This paper proposes a reasoning-aware hierarchical traversal mechanism that interleaves chain-of-thought generation at each depth-first search node evaluation step, incorporating the resulting reasoning embedding into a combined node scoring function alongside query similarity. The hierarchical tree is constructed through iterative chunking, embedding, clustering, and summarization; retrieval is then guided by this combined score under a dual-threshold pruning mechanism that adaptively controls traversal depth and breadth. The proposed approach is evaluated on the NarrativeQA and QuALITY benchmarks against a semantic similarity-based traversal baseline. On QuALITY, the method achieves marginal gains in overall accuracy (+0.7%) and F1 (+0.1%), while substantially reducing the abstain rate from 4.1% to 1.8%, with improvement concentrated on normal-difficulty questions; on hard questions, the baseline outperforms the proposed method. On NarrativeQA, ROUGE-L F1, BLEU-1, and BLEU-4 improve slightly, though METEOR decreases, reflecting inconsistent metric-level effects. These results suggest that step-wise CoT reasoning can improve node selection quality in hierarchical traversal, with gains most evident on standard-difficulty questions. Performance on hard questions and metric-level consistency across NarrativeQA remain open challenges, indicating that the quality of the generated reasoning signal is a key bottleneck for further improvement.

Zitieren

APA

Maassobirin, M. K., J.S., W. S., & Adziima, A. F. (2026). Improving Tree Traversal in Hierarchical RAG via Reasoning-Aware Approach with Interleaving Chain-of-Thought. bit-Tech, 9(1), 30-40. https://doi.org/10.32877/bt.v9i1.3785

Vancouver

Maassobirin MK, J.S. WS, Adziima AF. Improving Tree Traversal in Hierarchical RAG via Reasoning-Aware Approach with Interleaving Chain-of-Thought. bit-Tech. 2026;9(1):30-40. doi:10.32877/bt.v9i1.3785

BibTeX
@article{Maassobirin2026,
  author = {Maassobirin, Mukhamad Khafid and J.S., Wahyu Syaifullah and Adziima, Andri Fauzan},
  title = {Improving Tree Traversal in Hierarchical RAG via Reasoning-Aware Approach with Interleaving Chain-of-Thought},
  journal = {bit-Tech},
  year = {2026},
  volume = {9},
  number = {1},
  pages = {30--40},
  doi = {10.32877/bt.v9i1.3785}
}
Kostenlos starten →