At the 13th International Symposium on Information and Communication Technology (SOICT 2024), Nguyen Song Thien Long presented two groundbreaking works, namely “URAG: Implementing a Unified Hybrid RAG for Precise Answers in University Admission Chatbots – A Case Study at HCMUT” and “RAPID: Retrieval-Augmented Parallel Inference Drafting for Text-Based Video Event Retrieval”. URAG combines rule-based systems with Retrieval-Augmented Generation (RAG) to enhance chatbot accuracy and reliability. Successfully deployed at Ho Chi Minh City University of Technology (HCMUT), it has transformed how admission-related queries are addressed, providing students with precise and reliable answers. Meanwhile, RAPID introduces a novel approach to text-based video event retrieval, leveraging parallel query drafting and multimodal embeddings. This method efficiently extracts relevant events from large-scale video datasets, such as those containing hundreds of hours of footage, setting a new standard in multimedia retrieval. These works exemplify the potential of AI to address practical challenges in education and multimedia, offering innovative solutions that balance precision, efficiency, and scalability.
