# Multimodal Depression Detection on PDCD 2025 (val)

Source: [Multimodal Depression Detection on PDCD 2025 (val)](<https://www.sota2.com/research/sota/multimodal-depression-detection-on-pdcd-2025-val>)

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

Data updated: 2026-04-14T08:50:58.806Z

Evaluation of models for Multimodal Depression Detection using the validation split of the PDCD 2025 dataset, reporting standard classification metrics.

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 Depression Detection |
| Dataset (including any stated split) | PDCD 2025 (val) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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-depression-detection-on-pdcd-2025-val.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [F1 Score](<https://www.sota2.com/research/sota/multimodal-depression-detection-on-pdcd-2025-val.md?metric=F1%20Score>): Higher is better; unit: PERCENT.

- [Precision](<https://www.sota2.com/research/sota/multimodal-depression-detection-on-pdcd-2025-val.md?metric=Precision>): Higher is better; unit: PERCENT.

- [Recall](<https://www.sota2.com/research/sota/multimodal-depression-detection-on-pdcd-2025-val.md?metric=Recall>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| ACMG w Cross-modal (Qwen) | 81.25 | Not recorded | 69ddca327d0172140ec79b1f |
| ACMG w Unimodal (Qwen) | 80.88 | Not recorded | 69ddca327d0172140ec79b1f |
| Transformer(Qwen) | 79.78 | Not recorded | 69ddca327d0172140ec79b1f |
| ACMG w Cross-modal (RoBERTa) | 78.68 | Not recorded | 69ddca327d0172140ec79b1f |
| ACMG w Unimodal (RoBERTa) | 77.57 | Not recorded | 69ddca327d0172140ec79b1f |
| Transformer(RoBERTa) | 75.74 | Not recorded | 69ddca327d0172140ec79b1f |
| Gomez-Zaragoza et al. (2025) | 75.37 | Not recorded | 69ddca327d0172140ec79b1f |
| Al Hanai et al. (2018) | 72.43 | Not recorded | 69ddca327d0172140ec79b1f |

### ACMG w Cross-modal (Qwen)

Accuracy: 81.25. Result ID: 69ddfead47f34ce924c57cdf.

[Reporting paper](<https://www.sota2.com/research/paper/69ddca327d0172140ec79b1f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69ddca327d0172140ec79b1f.md>)

Recorded evaluation configuration: {"Pre-trained Language Model":"Qwen-embedding-0.6B","Gating configuration":"Cross-modal"}. This record does not assert independent reproduction.

### ACMG w Unimodal (Qwen)

Accuracy: 80.88. Result ID: 69ddfead47f34ce924c57cde.

[Reporting paper](<https://www.sota2.com/research/paper/69ddca327d0172140ec79b1f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69ddca327d0172140ec79b1f.md>)

Recorded evaluation configuration: {"Pre-trained Language Model":"Qwen-embedding-0.6B","Gating configuration":"Unimodal"}. This record does not assert independent reproduction.

### Transformer(Qwen)

Accuracy: 79.78. Result ID: 69ddfead47f34ce924c57cdd.

[Reporting paper](<https://www.sota2.com/research/paper/69ddca327d0172140ec79b1f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76591e209131cea6c86e.md>)

Recorded evaluation configuration: {"Pre-trained Language Model":"Qwen-embedding-0.6B"}. This record does not assert independent reproduction.

### ACMG w Cross-modal (RoBERTa)

Accuracy: 78.68. Result ID: 69ddfead47f34ce924c57cdc.

[Reporting paper](<https://www.sota2.com/research/paper/69ddca327d0172140ec79b1f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69ddca327d0172140ec79b1f.md>)

Recorded evaluation configuration: {"Pre-trained Language Model":"RoBERTa","Gating configuration":"Cross-modal"}. This record does not assert independent reproduction.

### ACMG w Unimodal (RoBERTa)

Accuracy: 77.57. Result ID: 69ddfead47f34ce924c57cdb.

[Reporting paper](<https://www.sota2.com/research/paper/69ddca327d0172140ec79b1f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69ddca327d0172140ec79b1f.md>)

Recorded evaluation configuration: {"Pre-trained Language Model":"RoBERTa","Gating configuration":"Unimodal"}. This record does not assert independent reproduction.

### Transformer(RoBERTa)

Accuracy: 75.74. Result ID: 69ddfead47f34ce924c57cda.

[Reporting paper](<https://www.sota2.com/research/paper/69ddca327d0172140ec79b1f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76591e209131cea6c86e.md>)

Recorded evaluation configuration: {"Pre-trained Language Model":"RoBERTa"}. This record does not assert independent reproduction.

### Gomez-Zaragoza et al. (2025)

Accuracy: 75.37. Result ID: 69ddfead47f34ce924c57cd9.

[Reporting paper](<https://www.sota2.com/research/paper/69ddca327d0172140ec79b1f.md>)

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

### Al Hanai et al. (2018)

Accuracy: 72.43. Result ID: 69ddfead47f34ce924c57cd8.

[Reporting paper](<https://www.sota2.com/research/paper/69ddca327d0172140ec79b1f.md>)

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

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