A cause fusion framework with information bottleneck for conversational causal emotion entailment

Conversational Causal Emotion Entailment (CEE) aims to identify the underlying causes or triggers of a particular emotion expressed within a dialogue. Conversations often involve multiple cause utterances for an emotion, making the interconnections between causes critical in addition to the emotion-cause links. Regrettably, existing approaches frequently overlook the inter-cause relationships. Furthermore, existing techniques predominantly emphasize causal-related features, failing to adequately consider cause-related features and address irrelevant emotional information. To tackle these challenges, we propose a novel framework called Cause Fusion framework with Information Bottleneck (CFIB), which takes into account the inter-cause relationships. CFIB also places greater emphasis on cause-related features and reduces the impact of irrelevant emotional information. Extensive experimental results demonstrate that our approach achieves state-of-the-art performance on two challenging test datasets, thus validating the effectiveness of our proposed model.