Distribution-based Emotion Recognition in Conversation
This work addresses emotion-aware conversational AI by handling label disagreements, but it is incremental as it builds on existing ERC methods with a focus on uncertainty.
The paper tackled emotion recognition in conversation by proposing a distribution-based framework that handles label ambiguity through uncertainty estimation, achieving improved classification accuracy and uncertainty estimation on the IEMOCAP dataset.
Automatic emotion recognition in conversation (ERC) is crucial for emotion-aware conversational artificial intelligence. This paper proposes a distribution-based framework that formulates ERC as a sequence-to-sequence problem for emotion distribution estimation. The inherent ambiguity of emotions and the subjectivity of human perception lead to disagreements in emotion labels, which is handled naturally in our framework from the perspective of uncertainty estimation in emotion distributions. A Bayesian training loss is introduced to improve the uncertainty estimation by conditioning each emotional state on an utterance-specific Dirichlet prior distribution. Experimental results on the IEMOCAP dataset show that ERC outperformed the single-utterance-based system, and the proposed distribution-based ERC methods have not only better classification accuracy, but also show improved uncertainty estimation.