E3mo-Bench (Expressed) - Open-Vocabulary Recognition (Emotion Wheel): leaderboard
E3mo-Bench (E³mo-Bench, 2026) expressed-emotion subset, about the emotion expressed by the people in a video: the model names the emotions freely, scored against the reference set with the OV-MER semantic emotion-wheel F-score in percent; higher is better.
Metric: Emotion-wheel set F-score (%; the OV-MER semantic emotion wheel alignment F-EW between the predicted and reference open-vocabulary emotion sets; 16 MLLMs on videos from eight emotion datasets, each assigned one affective perspective; subset: the emotion expressed by the people on screen). Source: arxiv.org. Saturation forecast: Around 2030. 16 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.4 | 57.73 | #123 |
| 2 | InternVL3-8B | 55.57 | #1688 |
| 3 | Qwen2.5-Omni-7B | 52.5 | #1789 |
| 4 | Claude Sonnet 4.6 | 52.04 | #161 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=e3mo-bench-expressed-open-vocabulary-recognition-emotion-wheel · How It Works · Data refreshed daily, snapshot 2026-11-13.