Year: 2025
Type: article
Abstract: The rapid expansion of scientific publications presents growing challenges for researchers who must locate, evaluate, and verify relevant literature across vast databases. To address these needs, Multi-Agent Systems (MAS) and Generative AI (GenAI) are being developed to automate research workflows spanning literature retrieval, summarization, and verification. This survey paper provides an overview of recent studies that integrate these technologies to support researchers. The paper analyses leading architectures, Retrieval-Augmented Generation (RAG) systems, verification tools, and recommendation mechanisms. Findings suggest that while progress is significant, full end-to-end automation remains limited. Future systems must prioritize transparency, explainability, and user-centred design to truly transform research practices
Institution: AISSMS Institute of Information Technology
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4416148729