ALOHA: Empowering Multilingual Agent for University Orientation with Hierarchical Retrieval

2025-05-13

AI Summary

To address the challenge faced by university faculty and students in accessing multilingual, timely, and accurate campus information during orientation, this paper proposes a domain-specific multilingual intelligent agent system. The system introduces a novel hierarchical retrieval architecture that integrates document- and paragraph-level retrieval-augmented generation (RAG), multilingual large language model (MLLM)-based understanding and generation, real-time external API invocation, and human-in-the-loop interaction. This design effectively mitigates key bottlenecks—including domain knowledge gaps, limited language coverage, and response latency. Human evaluation demonstrates statistically significant improvements over mainstream commercial chatbots and search engines in answer correctness, timeliness, and interaction usability. Deployed in practice, the system has served over 12,000 users, validating both its technical efficacy and real-world applicability.

Technology Category

• Multiagent Systems: Agent Communication
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Cognitive Modeling & Cognitive Systems: Agent Architectures

Application Category

• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

The rise of Large Language Models~(LLMs) revolutionizes information retrieval, allowing users to obtain required answers through complex instructions within conversations. However, publicly available services remain inadequate in addressing the needs of faculty and students to search campus-specific information. It is primarily due to the LLM's lack of domain-specific knowledge and the limitation of search engines in supporting multilingual and timely scenarios. To tackle these challenges, we introduce ALOHA, a multilingual agent enhanced by hierarchical retrieval for university orientation. We also integrate external APIs into the front-end interface to provide interactive service. The human evaluation and case study show our proposed system has strong capabilities to yield correct, timely, and user-friendly responses to the queries in multiple languages, surpassing commercial chatbots and search engines. The system has been deployed and has provided service for more than 12,000 people.

Problem

• Addressing LLMs' lack of domain-specific university knowledge
• Overcoming search engines' multilingual and timeliness limitations
• Providing accurate multilingual campus information retrieval

Innovation

• Hierarchical retrieval enhances multilingual university orientation
• Integration of external APIs for interactive front-end service
• Deployed system serving over 12,000 users effectively