ProReason: Multi-Modal Proactive Reasoning with Decoupled Eyesight and Wisdom

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhou, Jingqi, Wang, Sheng, Dong, Jingwei, Liu, Kai, Li, Lei, Gao, Jiahui, Jiang, Jiyue, Kong, Lingpeng, Wu, Chuan
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917016298848256
author Zhou, Jingqi
Wang, Sheng
Dong, Jingwei
Liu, Kai
Li, Lei
Gao, Jiahui
Jiang, Jiyue
Kong, Lingpeng
Wu, Chuan
author_facet Zhou, Jingqi
Wang, Sheng
Dong, Jingwei
Liu, Kai
Li, Lei
Gao, Jiahui
Jiang, Jiyue
Kong, Lingpeng
Wu, Chuan
contents Large vision-language models (LVLMs) have witnessed significant progress on visual understanding tasks. However, they often prioritize language knowledge over image information on visual reasoning tasks, incurring performance degradation. To tackle this issue, we first identify the drawbacks of existing solutions (i.e., limited multi-modal reasoning capacities, and insufficient and irrelevant visual descriptions). We then decompose visual reasoning process into two stages: proactive visual perception (i.e., eyesight) and textual reasoning (i.e., wisdom), and introduce a novel visual reasoning framework named ProReason. This framework features decoupled vision-reasoning capabilities and multi-run proactive perception. Briefly, given a multi-modal question, ProReason iterates proactive information collection and reasoning until the answer can be concluded with necessary and sufficient visual descriptions. Notably, the disassociation of capabilities allows seamless integration of existing large language models (LLMs) to compensate for the reasoning deficits of LVLMs. Our extensive experiments demonstrate that ProReason outperforms existing multi-step reasoning frameworks on various benchmarks for both open-source and closed-source models, with the average performance gain reaching 13.2%. Besides, the integration of LLMs allows ProReason to produce high-quality visual reasoning data, which empowers ProReason-distilled models (i.e., ProReason-VL and ProReason-Q3) to achieve superior performance in downstream tasks. Our insights into existing solutions and the decoupled perspective for feasible integration of LLMs illuminate future research on visual reasoning techniques, especially LLM-assisted ones.
format Preprint
id arxiv_https___arxiv_org_abs_2410_14138
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle ProReason: Multi-Modal Proactive Reasoning with Decoupled Eyesight and Wisdom
Zhou, Jingqi
Wang, Sheng
Dong, Jingwei
Liu, Kai
Li, Lei
Gao, Jiahui
Jiang, Jiyue
Kong, Lingpeng
Wu, Chuan
Computer Vision and Pattern Recognition
Artificial Intelligence
Large vision-language models (LVLMs) have witnessed significant progress on visual understanding tasks. However, they often prioritize language knowledge over image information on visual reasoning tasks, incurring performance degradation. To tackle this issue, we first identify the drawbacks of existing solutions (i.e., limited multi-modal reasoning capacities, and insufficient and irrelevant visual descriptions). We then decompose visual reasoning process into two stages: proactive visual perception (i.e., eyesight) and textual reasoning (i.e., wisdom), and introduce a novel visual reasoning framework named ProReason. This framework features decoupled vision-reasoning capabilities and multi-run proactive perception. Briefly, given a multi-modal question, ProReason iterates proactive information collection and reasoning until the answer can be concluded with necessary and sufficient visual descriptions. Notably, the disassociation of capabilities allows seamless integration of existing large language models (LLMs) to compensate for the reasoning deficits of LVLMs. Our extensive experiments demonstrate that ProReason outperforms existing multi-step reasoning frameworks on various benchmarks for both open-source and closed-source models, with the average performance gain reaching 13.2%. Besides, the integration of LLMs allows ProReason to produce high-quality visual reasoning data, which empowers ProReason-distilled models (i.e., ProReason-VL and ProReason-Q3) to achieve superior performance in downstream tasks. Our insights into existing solutions and the decoupled perspective for feasible integration of LLMs illuminate future research on visual reasoning techniques, especially LLM-assisted ones.
title ProReason: Multi-Modal Proactive Reasoning with Decoupled Eyesight and Wisdom
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2410.14138