Search papers, labs, and topics across Lattice.
This paper introduces a novel framework, TCDA, for Conversational Aspect-based Sentiment Quadruple Analysis (DiaASQ) that combines a Thread-Constrained Directed Acyclic Graph (TC-DAG) and Discourse-Aware Rotary Position Embedding (D-RoPE). TC-DAG filters structural noise and incorporates temporal dialogue sequences, while D-RoPE aligns multi-layer semantics and captures thread dependencies. Experiments on two benchmark datasets show that TCDA achieves state-of-the-art performance in capturing complex interrelationships in multi-round dialogues.
Achieve state-of-the-art conversational sentiment analysis by explicitly modeling discourse structure and thread dependencies, outperforming GCN-based approaches.
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.
