Saved in:
Bibliographic Details
Main Authors: Ma, Thanh, La, Tri-Tam, Huu, Lam-Thu Le, Nguyen, Minh-Nghi, Luu, Khanh-Van Pham
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2510.01800
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916071125024768
author Ma, Thanh
La, Tri-Tam
Huu, Lam-Thu Le
Nguyen, Minh-Nghi
Luu, Khanh-Van Pham
author_facet Ma, Thanh
La, Tri-Tam
Huu, Lam-Thu Le
Nguyen, Minh-Nghi
Luu, Khanh-Van Pham
contents Academic regulation advising is essential for helping students interpret and comply with institutional policies, yet building effective systems requires domain specific regulatory resources. To address this challenge, we propose REBot, an LLM enhanced advisory chatbot powered by CatRAG, a hybrid retrieval reasoning framework that integrates retrieval augmented generation with graph based reasoning. CatRAG unifies dense retrieval and graph reasoning, supported by a hierarchical, category labeled knowledge graph enriched with semantic features for domain alignment. A lightweight intent classifier routes queries to the appropriate retrieval modules, ensuring both factual accuracy and contextual depth. We construct a regulation specific dataset and evaluate REBot on classification and question answering tasks, achieving state of the art performance with an F1 score of 98.89%. Finally, we implement a web application that demonstrates the practical value of REBot in real world academic advising scenarios.
format Preprint
id arxiv_https___arxiv_org_abs_2510_01800
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle REBot: From RAG to CatRAG with Semantic Enrichment and Graph Routing
Ma, Thanh
La, Tri-Tam
Huu, Lam-Thu Le
Nguyen, Minh-Nghi
Luu, Khanh-Van Pham
Artificial Intelligence
Academic regulation advising is essential for helping students interpret and comply with institutional policies, yet building effective systems requires domain specific regulatory resources. To address this challenge, we propose REBot, an LLM enhanced advisory chatbot powered by CatRAG, a hybrid retrieval reasoning framework that integrates retrieval augmented generation with graph based reasoning. CatRAG unifies dense retrieval and graph reasoning, supported by a hierarchical, category labeled knowledge graph enriched with semantic features for domain alignment. A lightweight intent classifier routes queries to the appropriate retrieval modules, ensuring both factual accuracy and contextual depth. We construct a regulation specific dataset and evaluate REBot on classification and question answering tasks, achieving state of the art performance with an F1 score of 98.89%. Finally, we implement a web application that demonstrates the practical value of REBot in real world academic advising scenarios.
title REBot: From RAG to CatRAG with Semantic Enrichment and Graph Routing
topic Artificial Intelligence
url https://arxiv.org/abs/2510.01800