Less is More: A Simple yet Effective Token Reduction Method for Efficient Multi-modal LLMs
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_ | 1866916529366368256 |
|---|---|
| author | Song, Dingjie Wang, Wenjun Chen, Shunian Wang, Xidong Guan, Michael Wang, Benyou |
| author_facet | Song, Dingjie Wang, Wenjun Chen, Shunian Wang, Xidong Guan, Michael Wang, Benyou |
| contents | The rapid advancement of Multimodal Large Language Models (MLLMs) has led to remarkable performances across various domains. However, this progress is accompanied by a substantial surge in the resource consumption of these models. We address this pressing issue by introducing a new approach, Token Reduction using CLIP Metric (TRIM), aimed at improving the efficiency of MLLMs without sacrificing their performance. Inspired by human attention patterns in Visual Question Answering (VQA) tasks, TRIM presents a fresh perspective on the selection and reduction of image tokens. The TRIM method has been extensively tested across 12 datasets, and the results demonstrate a significant reduction in computational overhead while maintaining a consistent level of performance. This research marks a critical stride in efficient MLLM development, promoting greater accessibility and sustainability of high-performing models. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_10994 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Less is More: A Simple yet Effective Token Reduction Method for Efficient Multi-modal LLMs Song, Dingjie Wang, Wenjun Chen, Shunian Wang, Xidong Guan, Michael Wang, Benyou Computation and Language Artificial Intelligence Computer Vision and Pattern Recognition Multimedia The rapid advancement of Multimodal Large Language Models (MLLMs) has led to remarkable performances across various domains. However, this progress is accompanied by a substantial surge in the resource consumption of these models. We address this pressing issue by introducing a new approach, Token Reduction using CLIP Metric (TRIM), aimed at improving the efficiency of MLLMs without sacrificing their performance. Inspired by human attention patterns in Visual Question Answering (VQA) tasks, TRIM presents a fresh perspective on the selection and reduction of image tokens. The TRIM method has been extensively tested across 12 datasets, and the results demonstrate a significant reduction in computational overhead while maintaining a consistent level of performance. This research marks a critical stride in efficient MLLM development, promoting greater accessibility and sustainability of high-performing models. |
| title | Less is More: A Simple yet Effective Token Reduction Method for Efficient Multi-modal LLMs |
| topic | Computation and Language Artificial Intelligence Computer Vision and Pattern Recognition Multimedia |
| url | https://arxiv.org/abs/2409.10994 |