发表机构
Oklahoma State University（俄克拉荷马州立大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究针对社交媒体对话树的情感转变问题，提出C³T模型，基于CaSiRe数据集，联合预测节点情感与转变，在事件级划分下优于多类基线，为社交媒体分析提供结构导向的反事实建模动机。
AI 中文摘要
社交媒体帖子的情感不仅在不同帖子间存在差异，还会随用户对分支回复树中主张、修正、证据和敌意的反应而转变。本研究以话语动作（如否认/修正、证据/链接、毒性/攻击）作为候选干预措施，探讨谣言相关对话树中情感变化的原因，具体研究问题包括：回复表达何种情感、其情感是否相对于父节点发生转变、哪条先前消息最可能导致该回复的情感。为支撑该研究设置，我们推出CaSiRe——一个基于公开谣言对话数据集的因果情感推理层，添加了帖子级情感标签、诱导的父子情感转变标签、校准的多标签干预标签以及明确标注的因果源标签。随后，我们提出C³T（Counterfactual Causal Conversation Transformer，反事实因果对话Transformer），这是一种线程结构的时序模型，可联合预测节点情感与转变、学习稀疏祖先归因，并通过开启或关闭对话干预嵌入来支持反事实查询以估计潜在结果。在事件级划分下，C³T在事件外鲁棒性和归因方面优于仅文本、基于图和时序的基线模型，且产生可解释的模型效应：否认/修正和证据会降低下游负面情感，而毒性会增加下游负面情感。我们还对开放权重大语言模型（LLM）提示基线进行了基准测试，发现添加对话上下文有帮助，但归因仍不够可靠，这为社交媒体分析中基于结构的反事实建模提供了动机。
英文摘要
Sentiment in social-media threads does not only vary across posts; it shifts as users react to claims, corrections, evidence, and hostility within a branching reply tree. We study why sentiment changes in rumor-centric conversation trees by treating discourse moves (e.g., denial/correction, evidence/link, toxicity/attack) as candidate interventions and asking (i) what sentiment a reply expresses, (ii) whether the sentiment shifts relative to its parent, and (iii) which prior message most plausibly drove the reply's sentiment. To support this setting, we introduce CaSiRe, a causal sentiment reasoning layer over public rumor conversation datasets that adds post-level sentiment labels, induced parent-child shift labels, calibrated multi-label intervention tags, and explicitly annotated causal-source labels. We then propose C$^{3}$T (Counterfactual Causal Conversation Transformer), a thread-structured temporal model that jointly predicts node sentiment and shifts, learns sparse ancestor attribution, and supports counterfactual queries by forcing conversational intervention embeddings on or off to estimate potential outcomes. Under an event-level split, C$^{3}$T improves out-of-event robustness and attribution over text-only, graph-based, and temporal baselines, and yields interpretable model-based effects: denials/corrections and evidence reduce downstream negativity, while toxicity increases it. We also benchmark open-weight LLM prompting baselines and find that added conversational context helps, but attribution remains less reliable, motivating structure-aware counterfactual modeling for social-media analysis.
Comments23 pages, 3 figures, 7 tables; accepted to the EMNLP 2026 Main Conference
