CoLLaVO: Crayon Large Language and Vision mOdel
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_ | 1866913374940430336 |
|---|---|
| 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 remarkable success of Large Language Models (LLMs) and instruction tuning drives the evolution of Vision Language Models (VLMs) towards a versatile general-purpose model. Yet, it remains unexplored whether current VLMs genuinely possess quality object-level image understanding capabilities determined from 'what objects are in the image?' or 'which object corresponds to a specified bounding box?'. Our findings reveal that the image understanding capabilities of current VLMs are strongly correlated with their zero-shot performance on vision language (VL) tasks. This suggests that prioritizing basic image understanding is crucial for VLMs to excel at VL tasks. To enhance object-level image understanding, we propose Crayon Large Language and Vision mOdel (CoLLaVO), which incorporates instruction tuning with Crayon Prompt as a new visual prompt tuning scheme based on panoptic color maps. Furthermore, we present a learning strategy of Dual QLoRA to preserve object-level image understanding without forgetting it during visual instruction tuning, thereby achieving a significant leap in numerous VL benchmarks in a zero-shot setting. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_11248 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | CoLLaVO: Crayon Large Language and Vision mOdel Lee, Byung-Kwan Park, Beomchan Kim, Chae Won Ro, Yong Man Computer Vision and Pattern Recognition The remarkable success of Large Language Models (LLMs) and instruction tuning drives the evolution of Vision Language Models (VLMs) towards a versatile general-purpose model. Yet, it remains unexplored whether current VLMs genuinely possess quality object-level image understanding capabilities determined from 'what objects are in the image?' or 'which object corresponds to a specified bounding box?'. Our findings reveal that the image understanding capabilities of current VLMs are strongly correlated with their zero-shot performance on vision language (VL) tasks. This suggests that prioritizing basic image understanding is crucial for VLMs to excel at VL tasks. To enhance object-level image understanding, we propose Crayon Large Language and Vision mOdel (CoLLaVO), which incorporates instruction tuning with Crayon Prompt as a new visual prompt tuning scheme based on panoptic color maps. Furthermore, we present a learning strategy of Dual QLoRA to preserve object-level image understanding without forgetting it during visual instruction tuning, thereby achieving a significant leap in numerous VL benchmarks in a zero-shot setting. |
| title | CoLLaVO: Crayon Large Language and Vision mOdel |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2402.11248 |