MER 2026: Interlocutor Emotion, Fine-Grained Emotion, Emotion Preference, and Emotion Recognition from Physiological Signals

Zheng Lian, Jianhua Tao, Björn Schuller, Guoying Zhao, Erik Cambria

Keywords: affective computing

Abstract: With the recent advancements in embodied AI, enabling robots to better understand human emotions and enhance their emotional intelligence has emerged as a significant research focus. This development can facilitate their seamless integration into human society. Emotion is an internal human state expressed through various modalities, including audio, video, text, and physiological signals. This has spurred the growth of Multimodal Emotion Recognition (MER), which seeks to integrate diverse cues to achieve a more comprehensive understanding of human emotions. This year, we plan to launch MER 2026, the fourth edition in the MER series of challenges, designed to align with current research trends. Over the past three years, the MER series of challenges has grown into one of the largest emotion recognition challenges in the research community. It provides a common platform for researchers worldwide to explore recent trends and emerging tasks in the field. From MER2023 to MER2025, participant numbers steadily increased, and our dataset from last year was downloaded over 18,000 times on Hugging Face. This year, we will advertise more widely and aim to attract up to 200 teams. **We believe that the MER series of challenges will serve as a flagship challenge at ACM Multimedia.**

Email Sharing: We authorize the sharing of all author emails with Program Chairs.

Data Release: We authorize the release of our submission and author names to the public in the event of acceptance.

Submission Number: 6