MoAI: Mixture of All Intelligence for Large Language and Vision Models
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866929423421276160 |
|---|---|
| author | Lee, Byung-Kwan Park, Beomchan Kim, Chae Won Ro, Yong Man |
| author_facet | Lee, Byung-Kwan Park, Beomchan Kim, Chae Won Ro, Yong Man |
| contents | The rise of large language models (LLMs) and instruction tuning has led to the current trend of instruction-tuned large language and vision models (LLVMs). This trend involves either meticulously curating numerous instruction tuning datasets tailored to specific objectives or enlarging LLVMs to manage vast amounts of vision language (VL) data. However, current LLVMs have disregarded the detailed and comprehensive real-world scene understanding available from specialized computer vision (CV) models in visual perception tasks such as segmentation, detection, scene graph generation (SGG), and optical character recognition (OCR). Instead, the existing LLVMs rely mainly on the large capacity and emergent capabilities of their LLM backbones. Therefore, we present a new LLVM, Mixture of All Intelligence (MoAI), which leverages auxiliary visual information obtained from the outputs of external segmentation, detection, SGG, and OCR models. MoAI operates through two newly introduced modules: MoAI-Compressor and MoAI-Mixer. After verbalizing the outputs of the external CV models, the MoAI-Compressor aligns and condenses them to efficiently use relevant auxiliary visual information for VL tasks. MoAI-Mixer then blends three types of intelligence (1) visual features, (2) auxiliary features from the external CV models, and (3) language features by utilizing the concept of Mixture of Experts. Through this integration, MoAI significantly outperforms both open-source and closed-source LLVMs in numerous zero-shot VL tasks, particularly those related to real-world scene understanding such as object existence, positions, relations, and OCR without enlarging the model size or curating extra visual instruction tuning datasets. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_07508 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | MoAI: Mixture of All Intelligence for Large Language and Vision Models Lee, Byung-Kwan Park, Beomchan Kim, Chae Won Ro, Yong Man Computer Vision and Pattern Recognition The rise of large language models (LLMs) and instruction tuning has led to the current trend of instruction-tuned large language and vision models (LLVMs). This trend involves either meticulously curating numerous instruction tuning datasets tailored to specific objectives or enlarging LLVMs to manage vast amounts of vision language (VL) data. However, current LLVMs have disregarded the detailed and comprehensive real-world scene understanding available from specialized computer vision (CV) models in visual perception tasks such as segmentation, detection, scene graph generation (SGG), and optical character recognition (OCR). Instead, the existing LLVMs rely mainly on the large capacity and emergent capabilities of their LLM backbones. Therefore, we present a new LLVM, Mixture of All Intelligence (MoAI), which leverages auxiliary visual information obtained from the outputs of external segmentation, detection, SGG, and OCR models. MoAI operates through two newly introduced modules: MoAI-Compressor and MoAI-Mixer. After verbalizing the outputs of the external CV models, the MoAI-Compressor aligns and condenses them to efficiently use relevant auxiliary visual information for VL tasks. MoAI-Mixer then blends three types of intelligence (1) visual features, (2) auxiliary features from the external CV models, and (3) language features by utilizing the concept of Mixture of Experts. Through this integration, MoAI significantly outperforms both open-source and closed-source LLVMs in numerous zero-shot VL tasks, particularly those related to real-world scene understanding such as object existence, positions, relations, and OCR without enlarging the model size or curating extra visual instruction tuning datasets. |
| title | MoAI: Mixture of All Intelligence for Large Language and Vision Models |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2403.07508 |