Dissecting Medical RAG: Why Reranking Matters More Than Complexity in Question Answering
Black Sea Journal of Engineering and Science · 15.03.2026 · 9 (2) · 549-561
DOI: 10.34248/bsengineering.1849342 ↗Kurz gefasst
Fachartikel in der Zeitschrift Black Sea Journal of Engineering and Science, online erschienen am 15.03.2026. Verfasst von Hakan Emekci und 1 weiteren Autor:innen. Die Arbeit nennt 34 Literaturangaben.
Abstract (Originalsprache)
Retrieval-Augmented Generation (RAG) systems integrate large language models with information retrieval to ground responses in factual data. This study systematically evaluates the contribution of each RAG component in a medical question answering system through comprehensive ablation analysis. We designed a hierarchical RAG architecture with six key components: hierarchical intent classification, query rewriting, two-stage retrieval (dense retrieval with FAISS + cross-encoder reranking using Clinical-Longformer), and specialist routing. We conducted systematic ablation studies across seven configurations on 476 medical questions from MedQA benchmarks. Each configuration was evaluated independently using GPT-4o mini as an LLM judge across four metrics: context relevance, completeness, faithfulness, and correctness (1-5 Likert scale), with each metric assessed through separate evaluation calls to minimize inter-metric bias. Statistical significance was validated through paired t-tests with effect size calculations (Cohen’s d). The full system achieved an overall score of 3.64/5.0. Systematic ablation revealed two critical components: reranking (removal: -0.24 overall, P
Zitieren
APA
Emekci, H., & Roxas, D. Q. (2026). Dissecting Medical RAG: Why Reranking Matters More Than Complexity in Question Answering. Black Sea Journal of Engineering and Science, 9(2), 549-561. https://doi.org/10.34248/bsengineering.1849342
Vancouver
Emekci H, Roxas DQ. Dissecting Medical RAG: Why Reranking Matters More Than Complexity in Question Answering. Black Sea Journal of Engineering and Science. 2026;9(2):549-561. doi:10.34248/bsengineering.1849342
BibTeX
@article{Emekci2026,
author = {Emekci, Hakan and Roxas, Daniel Quillan},
title = {Dissecting Medical RAG: Why Reranking Matters More Than Complexity in Question Answering},
journal = {Black Sea Journal of Engineering and Science},
year = {2026},
volume = {9},
number = {2},
pages = {549--561},
doi = {10.34248/bsengineering.1849342}
}
EUVIMED