@inproceedings{kashif-etal-2026-emotion,
    title = "Emotion Recogniton in Conversations - empirical study",
    author = "Kashif, Rufaida  and
      Piwowarski, Benjamin  and
      Gomez Adorno, Helena",
    editor = "Bagdon, Christopher  and
      Vishnubhotla, Krishnapriya  and
      Lindquist, Kristen A.  and
      Ungar, Lyle  and
      Klinger, Roman  and
      Mohammad, Saif M.",
    booktitle = "Proceedings of Computational Affective Science ({CAS}) @ {LREC} 2026",
    month = may,
    year = "2026",
    address = "Palma, Mallorca (Spain)",
    publisher = "ELRA Language Resources Association (ELRA)",
    url = "https://preview.aclanthology.org/ingest-lrec/2026.cas-1.8/",
    doi = "10.63317/2ju47z39zdo4",
    pages = "93--104",
    abstract = "Emotion Recognition in Conversations (ERC) requires modeling complex contextual dependencies across dialog turns. While transformer-based models achieve strong performance on ERC benchmarks, several key design choices including context construction, optimization strategies, and imbalance handling remain insufficiently examined. In this work, we conduct a systematic empirical study of transformer-based ERC models across three benchmark datasets. We analyze the impact of context length and directionality, layer freezing, learning rate scheduling, parameter-efficient fine-tuning, and class imbalance mitigation strategies. Our results show that short-to-medium conversational context and moderate layer freezing provide stable and strong performance, while very long context windows, aggressive freezing, and parameter-efficient adaptation offer limited gains. Furthermore, imbalance-aware losses and data augmentation do not consistently outperform standard cross-entropy training. Overall, our findings provide practical insights into effective and stable design choices for transformer-based conversational emotion recognition."
}