Towards Semantic Equivalence of Tokenization in Multimodal LLM
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909511090962432 |
|---|---|
| author | Wu, Shengqiong Fei, Hao Li, Xiangtai Ji, Jiayi Zhang, Hanwang Chua, Tat-Seng Yan, Shuicheng |
| author_facet | Wu, Shengqiong Fei, Hao Li, Xiangtai Ji, Jiayi Zhang, Hanwang Chua, Tat-Seng Yan, Shuicheng |
| contents | Multimodal Large Language Models (MLLMs) have demonstrated exceptional capabilities in processing vision-language tasks. One of the crux of MLLMs lies in vision tokenization, which involves efficiently transforming input visual signals into feature representations that are most beneficial for LLMs. However, existing vision tokenizers, essential for semantic alignment between vision and language, remain problematic. Existing methods aggressively fragment visual input, corrupting the visual semantic integrity. To address this, this paper proposes a novel dynamic Semantic-Equivalent Vision Tokenizer (SeTok), which groups visual features into semantic units via a dynamic clustering algorithm, flexibly determining the number of tokens based on image complexity. The resulting vision tokens effectively preserve semantic integrity and capture both low-frequency and high-frequency visual features. The proposed MLLM (Setokim) equipped with SeTok significantly demonstrates superior performance across various tasks, as evidenced by our experimental results. The project page is at https://chocowu.github.io/SeTok-web/. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_05127 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Towards Semantic Equivalence of Tokenization in Multimodal LLM Wu, Shengqiong Fei, Hao Li, Xiangtai Ji, Jiayi Zhang, Hanwang Chua, Tat-Seng Yan, Shuicheng Computer Vision and Pattern Recognition Multimodal Large Language Models (MLLMs) have demonstrated exceptional capabilities in processing vision-language tasks. One of the crux of MLLMs lies in vision tokenization, which involves efficiently transforming input visual signals into feature representations that are most beneficial for LLMs. However, existing vision tokenizers, essential for semantic alignment between vision and language, remain problematic. Existing methods aggressively fragment visual input, corrupting the visual semantic integrity. To address this, this paper proposes a novel dynamic Semantic-Equivalent Vision Tokenizer (SeTok), which groups visual features into semantic units via a dynamic clustering algorithm, flexibly determining the number of tokens based on image complexity. The resulting vision tokens effectively preserve semantic integrity and capture both low-frequency and high-frequency visual features. The proposed MLLM (Setokim) equipped with SeTok significantly demonstrates superior performance across various tasks, as evidenced by our experimental results. The project page is at https://chocowu.github.io/SeTok-web/. |
| title | Towards Semantic Equivalence of Tokenization in Multimodal LLM |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2406.05127 |