A Modular, Data-Free Pipeline for Multi-Label Intention Recognition in Transportation Agentic AI Applications

2025-11-05

AI Summary

To address the challenges of heavy reliance on labeled data, limited support for fine-grained multi-label prediction, and high data acquisition costs in traffic-domain intent recognition, this paper proposes DMTC—a fully unsupervised modular framework. DMTC integrates prompt engineering to generate diverse traffic-related queries, Sentence-T5 for semantic embedding, a lightweight classifier, and an online focal contrastive (OFC) loss to enhance hard-sample learning and inter-class discriminability without manual annotations. Evaluated on maritime traffic scenarios, DMTC achieves a Hamming loss of 5.35%, an AUC of 95.92%, and improves subset accuracy by over 3.29%—significantly outperforming state-of-the-art multi-label and end-to-end LLM-based methods. The framework establishes a new, efficient, and scalable paradigm for intent understanding in low-resource traffic intelligent agents.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Planning, Routing, and Scheduling: Planning with Language Models
• Natural Language Processing: Sentence-level Semantics, Textual Inference, etc.

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

In this study, a modular, data-free pipeline for multi-label intention recognition is proposed for agentic AI applications in transportation. Unlike traditional intent recognition systems that depend on large, annotated corpora and often struggle with fine-grained, multi-label discrimination, our approach eliminates the need for costly data collection while enhancing the accuracy of multi-label intention understanding. Specifically, the overall pipeline, named DMTC, consists of three steps: 1) using prompt engineering to guide large language models (LLMs) to generate diverse synthetic queries in different transport scenarios; 2) encoding each textual query with a Sentence-T5 model to obtain compact semantic embeddings; 3) training a lightweight classifier using a novel online focal-contrastive (OFC) loss that emphasizes hard samples and maximizes inter-class separability. The applicability of the proposed pipeline is demonstrated in an agentic AI application in the maritime transportation context. Extensive experiments show that DMTC achieves a Hamming loss of 5.35% and an AUC of 95.92%, outperforming state-of-the-art multi-label classifiers and recent end-to-end SOTA LLM-based baselines. Further analysis reveals that Sentence-T5 embeddings improve subset accuracy by at least 3.29% over alternative encoders, and integrating the OFC loss yields an additional 0.98% gain compared to standard contrastive objectives. In conclusion, our system seamlessly routes user queries to task-specific modules (e.g., ETA information, traffic risk evaluation, and other typical scenarios in the transportation domain), laying the groundwork for fully autonomous, intention-aware agents without costly manual labelling.

Problem

• Eliminates dependency on annotated data for multi-label intent recognition
• Enhances fine-grained multi-label discrimination in transportation AI systems
• Overcomes limitations of traditional intent recognition requiring costly data collection

Innovation

• Generates synthetic queries via prompt engineering
• Encodes queries with Sentence-T5 embeddings
• Trains classifier with online focal-contrastive loss