Year: 2026
Type: article
Abstract: Retrieval-Augmented Generation (RAG) has emerged as a powerful paradigm for enhancing the factual accuracy andcontextual relevance of large language models by integrating external knowledge sources. This paper proposes aSummary-Based RAG system that incorporates automated web scraping to dynamically collect, process, and utilize upto-date information from online sources[1]. The system first employs web scraping techniques to extract raw textual datafrom relevant web pages in real time. This data is then preprocessed and condensed using advanced summarizationalgorithms to generate concise, information-rich representations.These summaries are stored in a structured retrievalindex, enabling efficient semantic search based on user queries[2]. Upon receiving a query, the system retrieves the mostrelevant summaries and augments the input prompt provided to the generative model. By leveraging summarized contentinstead of raw documents, the approach reduces noise, improves retrieval precision, and minimizes token usage, leadingto faster and more cost-effective responses.
Source: PRATIBODH
Language: English
Topic: Topic Modeling
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W7215206985
