Spatial-Aware Efficient Projector for MLLMs via Multi-Layer Feature Aggregation

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Qian, Shun, Liu, Bingquan, Sun, Chengjie, Xu, Zhen, Wang, Baoxun
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866913544898871296
author Qian, Shun
Liu, Bingquan
Sun, Chengjie
Xu, Zhen
Wang, Baoxun
author_facet Qian, Shun
Liu, Bingquan
Sun, Chengjie
Xu, Zhen
Wang, Baoxun
contents The projector plays a crucial role in multi-modal language models (MLLMs). The number of visual tokens it outputs affects the efficiency of the MLLM, while the quality of the visual tokens influences the visual understanding capabilities of the MLLM. Current explorations on the projector focus on reducing the number of visual tokens to improve efficiency, often overlooking the inherent spatial discrepancy between the serialized 2-dimensional visual token sequences and natural language token sequences. A Spatial-Aware Efficient Projector (SAEP) is proposed to address this issue. In detail, our SAEP method employs an modified separable depthwise convolution module on multi-layer visual features to enhance the spatial information of visual tokens. As a result, our SAEP method can not only largely reduce the number of visual tokens by 75\%, but also significantly improve the multimodal spatial understanding capability of MLLMs. Moreover, compared to existing projectors, our SAEP gets best performances on massive multimodal evaluation benchmarks, which denotes its effectiveness on bridging the modality gap.
format Preprint
id arxiv_https___arxiv_org_abs_2410_10319
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Spatial-Aware Efficient Projector for MLLMs via Multi-Layer Feature Aggregation
Qian, Shun
Liu, Bingquan
Sun, Chengjie
Xu, Zhen
Wang, Baoxun
Computer Vision and Pattern Recognition
Multimedia
The projector plays a crucial role in multi-modal language models (MLLMs). The number of visual tokens it outputs affects the efficiency of the MLLM, while the quality of the visual tokens influences the visual understanding capabilities of the MLLM. Current explorations on the projector focus on reducing the number of visual tokens to improve efficiency, often overlooking the inherent spatial discrepancy between the serialized 2-dimensional visual token sequences and natural language token sequences. A Spatial-Aware Efficient Projector (SAEP) is proposed to address this issue. In detail, our SAEP method employs an modified separable depthwise convolution module on multi-layer visual features to enhance the spatial information of visual tokens. As a result, our SAEP method can not only largely reduce the number of visual tokens by 75\%, but also significantly improve the multimodal spatial understanding capability of MLLMs. Moreover, compared to existing projectors, our SAEP gets best performances on massive multimodal evaluation benchmarks, which denotes its effectiveness on bridging the modality gap.
title Spatial-Aware Efficient Projector for MLLMs via Multi-Layer Feature Aggregation
topic Computer Vision and Pattern Recognition
Multimedia
url https://arxiv.org/abs/2410.10319