@inproceedings{tak-etal-2025-mechanistic,
    title = "Mechanistic Interpretability of Emotion Inference in Large Language Models",
    author = "Tak, Ala N.  and
      Banayeeanzade, Amin  and
      Bolourani, Anahita  and
      Kian, Mina  and
      Jia, Robin  and
      Gratch, Jonathan",
    editor = "Che, Wanxiang  and
      Nabende, Joyce  and
      Shutova, Ekaterina  and
      Pilehvar, Mohammad Taher",
    booktitle = "Findings of the Association for Computational Linguistics: ACL 2025",
    month = jul,
    year = "2025",
    address = "Vienna, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/landing_page/2025.findings-acl.679/",
    pages = "13090--13120",
    ISBN = "979-8-89176-256-5",
    abstract = "Large language models (LLMs) show promising capabilities in predicting human emotions from text. However, the mechanisms through which these models process emotional stimuli remain largely unexplored. Our study addresses this gap by investigating how autoregressive LLMs infer emotions, showing that emotion representations are functionally localized to specific regions in the model. Our evaluation includes diverse model families and sizes, and is supported by robustness checks. We then show that the identified representations are psychologically plausible by drawing on cognitive appraisal theory{---}a well-established psychological framework positing that emotions emerge from evaluations (appraisals) of environmental stimuli. By causally intervening on construed appraisal concepts, we steer the generation and show that the outputs align with theoretical and intuitive expectations. This work highlights a novel way to causally intervene and control emotion inference, potentially benefiting safety and alignment in sensitive affective domains."
}