Research

Paper

TESTING February 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

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.

Metadata

arXiv ID: 2602.17856
Provider: ARXIV
Primary Category: cs.IR
Published: 2026-02-19
Fetched: 2026-02-23 05:33

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2602.17856v1</id>\n    <title>Enhancing Scientific Literature Chatbots with Retrieval-Augmented Generation: A Performance Evaluation of Vector and Graph-Based Systems</title>\n    <updated>2026-02-19T21:42:02Z</updated>\n    <link href='https://arxiv.org/abs/2602.17856v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2602.17856v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>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.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.IR'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.AI'/>\n    <published>2026-02-19T21:42:02Z</published>\n    <arxiv:primary_category term='cs.IR'/>\n    <author>\n      <name>Hamideh Ghanadian</name>\n    </author>\n    <author>\n      <name>Amin Kamali</name>\n    </author>\n    <author>\n      <name>Mohammad Hossein Tekieh</name>\n    </author>\n  </entry>"
}