Transferring Textual Preferences to Vision-Language Understanding through Model Merging
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866909619326025728 |
|---|---|
| author | Li, Chen-An Lin, Tzu-Han Chen, Yun-Nung Lee, Hung-yi |
| author_facet | Li, Chen-An Lin, Tzu-Han Chen, Yun-Nung Lee, Hung-yi |
| contents | Large vision-language models (LVLMs) perform outstandingly across various multimodal tasks. However, their ability to evaluate generated content remains limited, and training vision-language reward models (VLRMs) with preference data is computationally expensive. This paper explores a training-free alternative by merging text-based reward models (RMs) with LVLMs to create VLRMs. Our approach shows that integrating these models leads to improved performance over LVLMs' scoring and text-based RMs, offering an efficient method for incorporating textual preferences into LVLMs. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_13487 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Transferring Textual Preferences to Vision-Language Understanding through Model Merging Li, Chen-An Lin, Tzu-Han Chen, Yun-Nung Lee, Hung-yi Computation and Language Artificial Intelligence Computer Vision and Pattern Recognition Machine Learning Large vision-language models (LVLMs) perform outstandingly across various multimodal tasks. However, their ability to evaluate generated content remains limited, and training vision-language reward models (VLRMs) with preference data is computationally expensive. This paper explores a training-free alternative by merging text-based reward models (RMs) with LVLMs to create VLRMs. Our approach shows that integrating these models leads to improved performance over LVLMs' scoring and text-based RMs, offering an efficient method for incorporating textual preferences into LVLMs. |
| title | Transferring Textual Preferences to Vision-Language Understanding through Model Merging |
| topic | Computation and Language Artificial Intelligence Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2502.13487 |