对话中的因果情绪识别：上下文饱和与话语标记证据
Causal Emotion Recognition in Conversation: Context Saturation and Discourse-Marker Evidence
AI总结：
通过系统消融实验发现对话上下文对情绪识别性能起主导作用但快速饱和，并揭示悲伤情绪与左边缘话语标记使用减少及更高上下文依赖性的关联。
摘要：
我们解决了对话情绪识别中两个长期存在的空白：哪些建模选择实质性地影响性能，以及识别结果如何与可解释的话语层面模式相关联。我们通过在IEMOCAP上进行系统研究并在MELD上进行跨数据集验证来研究这两个问题。对于识别，我们使用10个随机种子进行受控消融实验，并进行多重比较校正的配对显著性检验，得到三个发现。首先，对话上下文是主导因素，但性能快速饱和：大约90%的性能提升来自最近的前10-30轮对话，具体取决于标签集。其次，层级句子表示仅在仅话语设置中帮助最大，并在MELD上显示出明显优势，但一旦轮次级别的上下文可用，其益处消失，表明对话历史吸收了大量话语内部结构。第三，整合外部情感词典不会改善结果，这与预训练编码器已经捕获ERC所需的大部分情感信号一致。在严格因果设置下，我们的简单模型实现了强性能（4-way 82.69%；6-way加权F1 67.07%），表明无需未来轮次即可达到竞争性准确率。对于语言分析，我们检查了5,286个话语标记出现，发现情绪与标记位置之间存在可靠关联（p <.0001）。悲伤话语的左边缘标记使用率（21.9%）低于其他情绪（28-32%），这与左边缘标记与主动话语管理相关的观点一致。这与我们的识别结果一致，其中悲伤从对话上下文中获益最多（+22个百分点），表明悲伤可能比具有更强局部语用线索的情绪更依赖于上下文。
英文摘要：
We address two persistent gaps in Emotion Recognition in Conversation: which modeling choices materially affect performance, and how recognition findings connect to interpretable discourse-level patterns. We study both through a systematic investigation on IEMOCAP with cross-dataset validation on MELD. For recognition, we run controlled ablations with 10 random seeds and paired significance tests with multiple-comparisons correction, yielding three findings. First, conversational context is the dominant factor, but performance saturates quickly: roughly 90% of the gain is captured within the most recent 10-30 preceding turns, depending on the label set. Second, hierarchical sentence representations help most in utterance-only settings and show a clear advantage on MELD, but their benefit disappears once turn-level context is available, suggesting that conversational history subsumes much of the intra-utterance structure. Third, integrating an external affective lexicon does not improve results, consistent with pretrained encoders already capturing most of the affective signal needed for ERC. Under a strictly causal setting, our simple models achieve strong performance (82.69% 4-way; 67.07% 6-way weighted F1), showing that competitive accuracy is achievable without future turns. For linguistic analysis, we examine 5,286 discourse-marker occurrences and find a reliable association between emotion and marker position (p < .0001). Sad utterances show reduced left-periphery marker usage (21.9%) relative to other emotions (28-32%), consistent with accounts linking left-periphery markers to active discourse management. This aligns with our recognition results, where Sad benefits most from conversational context (+22 percentage points), suggesting sadness may be more context-dependent than emotions with stronger local pragmatic cues.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
