TCDA: Thread-Constrained Discourse-Aware Modeling for Conversational Sentiment Quadruple Analysis

2026-05-03

AI Summary

This work addresses key challenges in conversational emotion quadruple analysis—namely, difficulties in modeling multi-turn interactions, interference from structural noise, neglect of temporal dynamics, and distance-induced signal dilution—by proposing a Thread-Constrained Directed Acyclic Graph (TC-DAG) and a Utterance-aware Rotary Position Encoding (D-RoPE). TC-DAG effectively filters cross-thread noise while preserving global connectivity, and D-RoPE jointly captures utterance-level progression and token-level sequential order to mitigate long-range dependency decay. Integrated with a dual-stream projection framework and a multi-scale frequency alignment mechanism, the proposed model achieves state-of-the-art performance, significantly outperforming existing methods on two benchmark datasets.

Abstract

Conversational Aspect-based Sentiment Quadruple Analysis (DiaASQ) needs to capture the complex interrelationships in multiple rounds of dialogues. Existing methods usually employ simple Graph Convolutional Networks (GCN), which introduce structural noise and fail to consider the temporal sequence of the dialogues, or use standard RoPE, which implicitly captures relative distances in a flat sequence but cannot clearly separate the token-level syntactic order from the utterance-level progression, and may suffer from the Distance Dilution problem. To address these issues, we propose a new framework that combines Thread-Constrained Directed Acyclic Graph (TC-DAG) and Discourse-Aware Rotary Position Embedding (D-RoPE). Specifically, TC-DAG filters out cross-thread noise based on thread constraints, maintains global connectivity through root anchoring, and incorporates the temporal sequence of the dialogues. D-RoPE aligns multi-layer semantics using dual-stream projection and multi-scale frequency signals, captures thread dependencies using tree-like distances, and alleviates the token-level Distance Dilution problem by incorporating utterance-level progressions. Experimental results on two benchmark datasets demonstrate that our framework achieves state-of-the-art performance.

Problem

• Conversational Sentiment Analysis
• Aspect-based Sentiment Quadruple
• Discourse Structure
• Temporal Sequence
• Distance Dilution

Innovation

• Thread-Constrained DAG
• Discourse-Aware RoPE
• Conversational Sentiment Quadruple Analysis
• Distance Dilution
• Temporal Discourse Modeling