# Emotion Recognition on CMU-MOSEI (test) Source: [Emotion Recognition on CMU-MOSEI (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:34:03.641Z Evaluation of Emotion Recognition performance on the test split of the CMU-MOSEI dataset, reporting accuracy and F1 scores for specific emotions (e.g., Happy, Sad, Angry). A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Emotion Recognition | | Dataset (including any stated split) | CMU-MOSEI (test) | | Metric | Accuracy (7) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 41 of 41 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Accuracy (7)](): Higher is better; unit: PERCENT. - [Accuracy (Happy)](): Higher is better; unit: PERCENT. - [F1 Score (Happy)](): Higher is better; unit: PERCENT. - [Accuracy (Sad)](): Higher is better; unit: PERCENT. - [F1 Score (Sad)](): Higher is better; unit: PERCENT. - [Accuracy (Angry)](): Higher is better; unit: PERCENT. - [F1 Score (Angry)](): Higher is better; unit: PERCENT. - [Accuracy (Fear)](): Higher is better; unit: PERCENT. - [F1 Score (Fear)](): Higher is better; unit: PERCENT. - [Accuracy (Disgust)](): Higher is better; unit: PERCENT. - [F1 Score (Disgust)](): Higher is better; unit: PERCENT. - [Accuracy (Surprise)](): Higher is better; unit: PERCENT. - [F1 Score (Surprise)](): Higher is better; unit: PERCENT. - [F1 Score (Happiness, weighted)](): Higher is better; unit: PERCENT. - [F1 Score (Sadness, weighted)](): Higher is better; unit: PERCENT. - [F1 Score (Angry, weighted)](): Higher is better; unit: PERCENT. - [F1 Score (Fear, weighted)](): Higher is better; unit: PERCENT. - [F1 Score (Disgust, weighted)](): Higher is better; unit: PERCENT. - [F1 Score (Surprise, weighted)](): Higher is better; unit: PERCENT. - [Accuracy (2)](): Higher is better; unit: PERCENT. - [F1 Score (Overall)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy (7) | Result date | Report paper | | --- | --- | --- | --- | | DMD | 54.6 | Not recorded | 68fa6ee61e209131cea42bde | | DMD | 54.5 | Not recorded | 68fa6ee61e209131cea42bde | | FDMER | 54.1 | Not recorded | 68fa6ee61e209131cea42bde | | DMD | 53.7 | Not recorded | 68fa6ee61e209131cea42bde | | PMR | 52.5 | Not recorded | 68fa6ee61e209131cea42bde | | MICA | 52.4 | Not recorded | 68fa6ee61e209131cea42bde | | MISA | 52.2 | Not recorded | 68fa6ee61e209131cea42bde | | MulT | 51.8 | Not recorded | 68fa6ee61e209131cea42bde | | PMR | 51.8 | Not recorded | 68fa6ee61e209131cea42bde | | MulT | 50.7 | Not recorded | 68fa6ee61e209131cea42bde | | RAVEN | 50 | Not recorded | 68fa6ee61e209131cea42bde | | MCTN | 49.6 | Not recorded | 68fa6ee61e209131cea42bde | | LF-LSTM | 48.8 | Not recorded | 68fa6ee61e209131cea42bde | | LF-LSTM | 48.8 | Not recorded | 68fa6ee61e209131cea42bde | | MCTN | 48.2 | Not recorded | 68fa6ee61e209131cea42bde | | EF-LSTM | 47.4 | Not recorded | 68fa6ee61e209131cea42bde | | EF-LSTM | 46.3 | Not recorded | 68fa6ee61e209131cea42bde | | RAVEN | 45.5 | Not recorded | 68fa6ee61e209131cea42bde | | Graph-MFN | 45 | Not recorded | 68fa6ee61e209131cea42bde | ### DMD Accuracy (7): 54.6. Result ID: 69906ada3a6c314849006b0d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Setting":"Unaligned","BERT-based language features":false}. This record does not assert independent reproduction. ### DMD Accuracy (7): 54.5. Result ID: 69906ada3a6c314849006b05. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Setting":"Aligned","BERT-based language features":true}. This record does not assert independent reproduction. ### FDMER Accuracy (7): 54.1. Result ID: 69906ada3a6c314849006b04. [Reporting paper]() Recorded evaluation configuration: {"Setting":"Aligned","BERT-based language features":true}. This record does not assert independent reproduction. ### DMD Accuracy (7): 53.7. Result ID: 69906ada3a6c314849006b02. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Setting":"Aligned","BERT-based language features":false}. This record does not assert independent reproduction. ### PMR Accuracy (7): 52.5. Result ID: 69906ada3a6c314849006b01. [Reporting paper]() Recorded evaluation configuration: {"Setting":"Aligned","BERT-based language features":false}. This record does not assert independent reproduction. ### MICA Accuracy (7): 52.4. Result ID: 69906ada3a6c314849006b0c. [Reporting paper]() Recorded evaluation configuration: {"Setting":"Unaligned","BERT-based language features":false}. This record does not assert independent reproduction. ### MISA Accuracy (7): 52.2. Result ID: 69906ada3a6c314849006b03. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Setting":"Aligned","BERT-based language features":true}. This record does not assert independent reproduction. ### MulT Accuracy (7): 51.8. Result ID: 69906ada3a6c314849006b00. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Setting":"Aligned","BERT-based language features":false}. This record does not assert independent reproduction. ### PMR Accuracy (7): 51.8. Result ID: 69906ada3a6c314849006b0b. [Reporting paper]() Recorded evaluation configuration: {"Setting":"Unaligned","BERT-based language features":false}. This record does not assert independent reproduction. ### MulT Accuracy (7): 50.7. Result ID: 69906ada3a6c314849006b0a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Setting":"Unaligned","BERT-based language features":false}. This record does not assert independent reproduction. ### RAVEN Accuracy (7): 50. Result ID: 69906ada3a6c314849006afe. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Setting":"Aligned","BERT-based language features":false}. This record does not assert independent reproduction. ### MCTN Accuracy (7): 49.6. Result ID: 69906ada3a6c314849006aff. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Setting":"Aligned","BERT-based language features":false}. This record does not assert independent reproduction. ### LF-LSTM Accuracy (7): 48.8. Result ID: 69906ada3a6c314849006afc. [Reporting paper]() Recorded evaluation configuration: {"Setting":"Aligned","BERT-based language features":false}. This record does not assert independent reproduction. ### LF-LSTM Accuracy (7): 48.8. Result ID: 69906ada3a6c314849006b07. [Reporting paper]() Recorded evaluation configuration: {"Setting":"Unaligned","BERT-based language features":false}. This record does not assert independent reproduction. ### MCTN Accuracy (7): 48.2. Result ID: 69906ada3a6c314849006b09. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Setting":"Unaligned","BERT-based language features":false}. This record does not assert independent reproduction. ### EF-LSTM Accuracy (7): 47.4. Result ID: 69906ada3a6c314849006afb. [Reporting paper]() Recorded evaluation configuration: {"Setting":"Aligned","BERT-based language features":false}. This record does not assert independent reproduction. ### EF-LSTM Accuracy (7): 46.3. Result ID: 69906ada3a6c314849006b06. [Reporting paper]() Recorded evaluation configuration: {"Setting":"Unaligned","BERT-based language features":false}. This record does not assert independent reproduction. ### RAVEN Accuracy (7): 45.5. Result ID: 69906ada3a6c314849006b08. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Setting":"Unaligned","BERT-based language features":false}. This record does not assert independent reproduction. ### Graph-MFN Accuracy (7): 45. Result ID: 69906ada3a6c314849006afd. [Reporting paper]() Recorded evaluation configuration: {"Setting":"Aligned","BERT-based language features":false}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.