DASH: Dialogue-Aware Similarity and Handshake Recognition for Topic Segmentation in Public-Channel Conversations

2025-12-16

AI Summary

Dialogue Topic Segmentation (DTS) faces significant challenges in informal public-channel dialogues—such as maritime VHF communications—including implicit topic shifts, syntactically loose language, and ambiguous transitions. To address these, this paper proposes a fine-grained DTS framework tailored to real-world maritime scenarios. It introduces the novel “dialogue handshake identification” mechanism to explicitly model topic-transition signals; designs a similarity-driven contextual exemplar selection strategy coupled with selective positive/negative sample augmentation to enhance discriminative robustness; and leverages large language models (LLMs) for semantic modeling, dynamic sample generation, confidence calibration, and interpretable reasoning. We present VHF-Dial—the first publicly available, real-world maritime VHF dialogue dataset—open-sourced to advance research. Our method achieves state-of-the-art performance on VHF-Dial and multiple benchmark datasets, enabling segment-level explainable segmentation and real-time confidence estimation, thereby establishing a new paradigm for maritime communication monitoring and decision support.

Technology Category

• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Natural Language Processing: Conversational AI/Dialog Systems
• Multiagent Systems: Agent Communication

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Dialogue Topic Segmentation (DTS) is crucial for understanding task-oriented public-channel communications, such as maritime VHF dialogues, which feature informal speech and implicit transitions. To address the limitations of traditional methods, we propose DASH-DTS, a novel LLM-based framework. Its core contributions are: (1) topic shift detection via dialogue handshake recognition; (2) contextual enhancement through similarity-guided example selection; and (3) the generation of selective positive and negative samples to improve model discrimination and robustness. Additionally, we release VHF-Dial, the first public dataset of real-world maritime VHF communications, to advance research in this domain. DASH-DTS provides interpretable reasoning and confidence scores for each segment. Experimental results demonstrate that our framework achieves several sota segmentation trusted accuracy on both VHF-Dial and standard benchmarks, establishing a strong foundation for stable monitoring and decision support in operational dialogues.

Problem

• Segmenting topics in noisy public-channel dialogues
• Detecting implicit topic shifts via handshake recognition
• Enhancing segmentation with similarity-guided context selection

Innovation

• Topic shift detection via dialogue handshake recognition
• Contextual enhancement through similarity-guided example selection
• Generation of selective positive and negative samples