Enhancing Scientific Literature Chatbots with Retrieval-Augmented Generation: A Performance Evaluation of Vector and Graph-Based Systems

Hamideh Ghanadian; Amin Kamali; Mohammad Hossein Tekieh

cs.IRFeb 19, 2026

Enhancing Scientific Literature Chatbots with Retrieval-Augmented Generation: A Performance Evaluation of Vector and Graph-Based Systems

Authors: Hamideh Ghanadian, Amin Kamali, Mohammad Hossein Tekieh

Organizations: Pest Management Regulatory Agency (PMRA), Health Canada, Ottawa, ON, Canada · Faculty of Engineering, University of Ottawa, Ottawa, ON, Canada · Oral Health Branch (OHB), Health Canada, Ottawa, ON, Canada

Abstract

This paper investigates the enhancement of scientific literature chatbots through retrieval-augmented generation (RAG), with a focus on evaluating vector- and graph-based retrieval systems. The proposed chatbot leverages both structured (graph) and unstructured (vector) databases to access scientific articles and gray literature, enabling efficient triage of sources according to research objectives. To systematically assess performance, we examine two use-case scenarios: retrieval from a single uploaded document and retrieval from a large-scale corpus. Benchmark test sets were generated using a GPT model, with selected outputs annotated for evaluation. The comparative analysis emphasizes retrieval accuracy and response relevance, providing insight into the strengths and limitations of each approach. The findings demonstrate the potential of hybrid RAG systems to improve accessibility to scientific knowledge and to support evidence-based decision making.

Figures & tables

Figure 1. Workflow of a Vector-based Retrieval-Augmented Generation

Test Dataset	Is the Context related to the Question?	Does the Answer fully answer the Question?
Single-Paper Test	94.2%	92%
Multi-Paper Test	81.6%	85%

Table 1. Evaluation of the quality of two synthetic test sets generated by LLM

	RAG on Single Paper				RAG on Database
Methods/Metrics	Cosine Similarity		Faithfulness		Cosine Similarity		Faithfulness
	Mean	SD.	Mean	SD.	Mean	SD.	Mean	SD.
VectorRAG	0.670	0.189	0.841	0.137	0.651	0.188	0.830	0.133
GraphRAG	0.654	0.184	0.785	0.153	0.633	0.174	0.761	0.149
Hybrid RAG	0.687	0.187	0.845	0.119	0.662	0.196	0.834	0.121

Table 2. Evaluation of the RAG performance on the two synthetic test datasets

RAG Evaluation

Retrieval-augmented generation (RAG) evaluation measures how well systems retrieve, use, and cite evidence while balancing reliability, cost, and speed.

See all 353 papers

VecTree-RAG: An Agentic Retrieval-Augmented Generation Framework Combining Vector and Tree Retrieval for Efficiency and Accuracy
Jul 25, 2026Xinyan Zhong, Yuwei Shi, Yuqi Wei +3Scientific QA Agentic RAG
Department of Chemistry and Materials Science, Xi’an Jiaotong-Liverpool University, Suzhou 215123, Jiangsu, P. R. China · Suzhou Lab, Suzhou, P. R. China · College of Carbon Neutrality Future Technology, State Key Laboratory of Heavy Oil Processing, China University of Petroleum (Beijing), Beijing 102249, China +1
Re-ranking and Late Interaction Drive Retrieval Quality: A Controlled Comparison of RAG Strategies for Scientific Question Answering
Sep 29, 2026Bhagyesh Rathi, Eshan Chawla, William B. AndreopoulosLate Interaction Retrieval Scientific QA
San Jose State University, Department of Computer Science
Towards Robust Retrieval-Augmented Generation Based on Knowledge Graph: A Comparative Analysis
Mar 5, 2026Hazem Amamou, Stéphane Gagnon, Alan Davoust +1RAG Robustness Graph Retrieval-Augmented Generation
Institut national de la recherche scientifique (INRS-EMT), Montréal, Québec, Canada · INRS-UQO Mixed Research Unit on Cybersecurity, Gatineau, Québec, Canada · Université du Québec en Outaouais, Gatineau, Québec, Canada

Jul 25, 2026J/K move · Enter open · S save

VecTree-RAG: An Agentic Retrieval-Augmented Generation Framework Combining Vector and Tree Retrieval for Efficiency and Accuracy

Xinyan Zhong, Yuwei Shi, Yuqi Wei, Chen Shen+2

Department of Chemistry and Materials Science, Xi’an Jiaotong-Liverpool University, Suzhou 215123, Jiangsu, P. R. China · Suzhou Lab, Suzhou, P. R. China · College of Carbon Neutrality Future Technology, State Key Laboratory of Heavy Oil Processing, China University of Petroleum (Beijing), Beijing 102249, China+1

Scientific QA Agentic RAG Information Retrieval Hybrid Retrieval Evidence Retrieval

Open paper AskPDF arXiv

Scientific question answering requires a retrieval system to solve two distinct problems: identifying which papers are relevant and locating the supporting evidence within those papers. Conventional retrieval-augmented generation typically addresses both through similarity search over fixed-length passages, flattening document structure and separating scientific claims from their methodological and argumentative context. We present VecTree-RAG, an agentic framework that assigns these tasks to complementary retrieval mechanisms. Vector search ranks compact document and section representations across the corpus, whereas reasoning-guided traversal of source-verified section trees localizes evidence within shortlisted papers. Full text is retained in a page store and exposed progressively only after structural localization. We evaluate VecTree-RAG on 300 QASPER questions, an open-access subset of 54 LitQA2 questions, and 49 multi-document MOSAIC questions. Compared with Dense RAG, reranked Dense RAG, RAPTOR, and Search-o1, VecTree-RAG obtained the highest observed answer score on all three benchmarks, reaching 0.800 LLM-judge correctness on QASPER, 0.925 accuracy on LitQA2, and a 0.547 composite score on MOSAIC. On QASPER, its evidence-page precision was 0.274, compared with 0.046--0.071 for the baselines. LitQA2 ablations further showed that the complete vector--tree architecture required fewer inference tokens than variants without tree navigation or corpus-level vector routing. These results indicate that vector retrieval narrows the corpus-level search space and tree navigation concentrates reading on structurally relevant evidence. Although multi-turn inference remains more expensive than single-call retrieval, VecTree-RAG provides a structure-aware and traceable architecture for scientific literature question answering.

Abstract

RAG Evaluation

Explore similar work

VecTree-RAG: An Agentic Retrieval-Augmented Generation Framework Combining Vector and Tree Retrieval for Efficiency and Accuracy

Re-ranking and Late Interaction Drive Retrieval Quality: A Controlled Comparison of RAG Strategies for Scientific Question Answering

Towards Robust Retrieval-Augmented Generation Based on Knowledge Graph: A Comparative Analysis