REIC: RAG-Enhanced Intent Classification at Scale

2025-05-30

AI Summary

To address scalability bottlenecks in intent classification for large-scale customer service platforms—specifically, the rapid growth in intent vocabulary and heterogeneous classification schemas across domains—this paper proposes a retrieval-augmented dynamic intent recognition framework. The method deeply integrates Retrieval-Augmented Generation (RAG) into the intent classification pipeline: semantic retrieval retrieves relevant external knowledge, which is then dynamically injected via large language model (LLM) prompt engineering; additionally, a lightweight intent adapter is designed to jointly optimize in-domain generalization and cross-domain transferability. Evaluated on real-world customer service datasets, the approach significantly outperforms fine-tuning, zero-shot, and few-shot baselines: it achieves a 12.3% absolute gain in in-domain accuracy and an F1 score of 86.7% in cross-domain scenarios, while eliminating the need for frequent retraining—thereby enabling agile product-line expansion.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Natural Language Processing: Safety and Robustness

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Accurate intent classification is critical for efficient routing in customer service, ensuring customers are connected with the most suitable agents while reducing handling times and operational costs. However, as companies expand their product lines, intent classification faces scalability challenges due to the increasing number of intents and variations in taxonomy across different verticals. In this paper, we introduce REIC, a Retrieval-augmented generation Enhanced Intent Classification approach, which addresses these challenges effectively. REIC leverages retrieval-augmented generation (RAG) to dynamically incorporate relevant knowledge, enabling precise classification without the need for frequent retraining. Through extensive experiments on real-world datasets, we demonstrate that REIC outperforms traditional fine-tuning, zero-shot, and few-shot methods in large-scale customer service settings. Our results highlight its effectiveness in both in-domain and out-of-domain scenarios, demonstrating its potential for real-world deployment in adaptive and large-scale intent classification systems.

Problem

• Scalability challenges in intent classification due to increasing intents
• Need for precise classification without frequent retraining
• Improving performance in large-scale customer service settings

Innovation

• Retrieval-augmented generation for dynamic knowledge
• No frequent retraining for precise classification
• Outperforms traditional methods in large-scale settings