Attention Isn't All You Need for Emotion Recognition:Domain Features Outperform Transformers on the EAV Dataset
Fuente:
arXiv
Guardado en:
| Autor principal: | |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866910008304730112 |
|---|---|
| author | Guragain, Anmol |
| author_facet | Guragain, Anmol |
| contents | We present a systematic study of multimodal emotion recognition using the EAV dataset, investigating whether complex attention mechanisms improve performance on small datasets. We implement three model categories: baseline transformers (M1), novel factorized attention mechanisms (M2), and improved CNN baselines (M3). Our experiments show that sophisticated attention mechanisms consistently underperform on small datasets. M2 models achieved 5 to 13 percentage points below baselines due to overfitting and destruction of pretrained features. In contrast, simple domain-appropriate modifications proved effective: adding delta MFCCs to the audio CNN improved accuracy from 61.9% to 65.56% (+3.66pp), while frequency-domain features for EEG achieved 67.62% (+7.62pp over the paper baseline). Our vision transformer baseline (M1) reached 75.30%, exceeding the paper's ViViT result (74.5%) through domain-specific pretraining, and vision delta features achieved 72.68% (+1.28pp over the paper CNN). These findings demonstrate that for small-scale emotion recognition, domain knowledge and proper implementation outperform architectural complexity. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_22161 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Attention Isn't All You Need for Emotion Recognition:Domain Features Outperform Transformers on the EAV Dataset Guragain, Anmol Machine Learning Computer Vision and Pattern Recognition Sound Audio and Speech Processing We present a systematic study of multimodal emotion recognition using the EAV dataset, investigating whether complex attention mechanisms improve performance on small datasets. We implement three model categories: baseline transformers (M1), novel factorized attention mechanisms (M2), and improved CNN baselines (M3). Our experiments show that sophisticated attention mechanisms consistently underperform on small datasets. M2 models achieved 5 to 13 percentage points below baselines due to overfitting and destruction of pretrained features. In contrast, simple domain-appropriate modifications proved effective: adding delta MFCCs to the audio CNN improved accuracy from 61.9% to 65.56% (+3.66pp), while frequency-domain features for EEG achieved 67.62% (+7.62pp over the paper baseline). Our vision transformer baseline (M1) reached 75.30%, exceeding the paper's ViViT result (74.5%) through domain-specific pretraining, and vision delta features achieved 72.68% (+1.28pp over the paper CNN). These findings demonstrate that for small-scale emotion recognition, domain knowledge and proper implementation outperform architectural complexity. |
| title | Attention Isn't All You Need for Emotion Recognition:Domain Features Outperform Transformers on the EAV Dataset |
| topic | Machine Learning Computer Vision and Pattern Recognition Sound Audio and Speech Processing |
| url | https://arxiv.org/abs/2601.22161 |