# Multimodal Emotion Recognition on CMU-MOSEI

Source: [Multimodal Emotion Recognition on CMU-MOSEI](<https://www.sota2.com/research/sota/multimodal-emotion-recognition-on-cmu-mosei>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-06-09T06:51:41.444Z

Evaluation of Multimodal Emotion Recognition performance on the CMU-MOSEI dataset, reporting Accuracy, Weighted F1, and Macro F1 scores.

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 | Multimodal Emotion Recognition |
| Dataset (including any stated split) | CMU-MOSEI |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| 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](<https://www.sota2.com/research/sota/multimodal-emotion-recognition-on-cmu-mosei.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [Weighted F1](<https://www.sota2.com/research/sota/multimodal-emotion-recognition-on-cmu-mosei.md?metric=Weighted%20F1>): Higher is better; unit: PERCENT.

- [Macro F1](<https://www.sota2.com/research/sota/multimodal-emotion-recognition-on-cmu-mosei.md?metric=Macro%20F1>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Segment Level Attention + Bi-Modal Transformer | 76.81 | Not recorded | 6a279e0b90966ade919bf668 |
| Memory Bear AI | 66.7 | Not recorded | 69c36c180636e5bf4afc650d |
| MulT | 65.4 | Not recorded | 69c36c180636e5bf4afc650d |
| EmoEmbs | 64.2 | Not recorded | 69c36c180636e5bf4afc650d |
| LF-LSTM | 63.1 | Not recorded | 69c36c180636e5bf4afc650d |

### Segment Level Attention + Bi-Modal Transformer

Accuracy: 76.81. Result ID: 6a27b7eefbc373592c23e459.

[Reporting paper](<https://www.sota2.com/research/paper/6a279e0b90966ade919bf668.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Memory Bear AI

Accuracy: 66.7. Result ID: 69c3d82e47f34ce924ad1ae5.

[Reporting paper](<https://www.sota2.com/research/paper/69c36c180636e5bf4afc650d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69c36c180636e5bf4afc650d.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### MulT

Accuracy: 65.4. Result ID: 69c3d82e47f34ce924ad1ae3.

[Reporting paper](<https://www.sota2.com/research/paper/69c36c180636e5bf4afc650d.md>)

Recorded evaluation configuration: {"Description":"Strong neural multimodal baseline"}. This record does not assert independent reproduction.

### EmoEmbs

Accuracy: 64.2. Result ID: 69c3d82e47f34ce924ad1ae4.

[Reporting paper](<https://www.sota2.com/research/paper/69c36c180636e5bf4afc650d.md>)

Recorded evaluation configuration: {"Description":"Context-/emotion-aware baseline"}. This record does not assert independent reproduction.

### LF-LSTM

Accuracy: 63.1. Result ID: 69c3d82e47f34ce924ad1ae2.

[Reporting paper](<https://www.sota2.com/research/paper/69c36c180636e5bf4afc650d.md>)

Recorded evaluation configuration: {"Description":"Traditional fusion baseline"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
