发表机构
The Chinese University of Hong Kong; The Hong Kong Polytechnic University; Harbin Institute of Technology（香港中文大学; 香港理工大学; 哈尔滨工业大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对对话式立场检测中历史对话信息利用的噪声问题，提出目标感知记忆图TamGraph，通过熵引导回溯机制动态构建目标感知图，提升LLM在该任务上的性能。
AI 中文摘要
立场检测对于理解表达对目标的潜在态度至关重要。对话式立场检测是真实社交媒体场景中更具挑战性的立场检测任务，因为它需要利用对话会话中与目标相关的历史语句来检测用户的立场。本文提出了目标感知记忆图TamGraph这一新方法，用于对话式立场检测，该方法动态利用与目标相关的语句。与立场检测时要么考虑所有先前历史对话、要么不使用任何先前对话信息的做法不同，TamGraph采用逐步的、熵引导的回溯机制，有选择地激活历史对话中的记忆，并动态构建目标感知图以建模话语间的立场关系。这使得既能利用对话历史中与目标相关的信息进行立场检测，又能防止引入噪声。在英文和中文基准上的实验结果表明，TamGraph显著提升了大型语言模型（LLM）在对话式立场检测任务上的性能。
英文摘要
Stance detection is crucial for understanding the underlying attitude of an expression towards a target. Conversational stance detection is a more challenging stance detection task in real-world social media scenarios, as it involves detecting the user's stance by leveraging the target-related historical statements across conversational sessions. In this paper, we propose target-aware Memory Graph TamGraph, a novel method that dynamically leverages target-related statements for conversational stance detection. Instead of considering all preceding historical conversations or using no prior conversation information for stance detection, our TamGraph employs a stepwise, entropy-guided backtracking mechanism to selectively activate memory from historical conversations and dynamically constructs a target-aware graph to model the stance relations among utterances. This allows the exploitation of target-related information from the conversation history for stance detection while preventing the introduction of noise. Experimental results on both English and Chinese benchmarks demonstrate that our TamGraph substantially improves LLM performance on conversational stance detection.
CommentsAccepted in EMNLP 2026 main
