构建多领域视觉推理数据集，训练出通用视觉语言模型Vision-G1
Vision-G1: Towards General Vision Language Reasoning with Multi-Domain Data Curation
- 从46个数据源筛选8类任务数据，用影响函数选优质样本
- 通过多轮强化学习与数据课程训练，模型在多个评测中领先
- 适合研究多模态推理、数据构建或希望超越闭源模型的开发者
尽管现有视觉语言模型在数学和逻辑推理等特定任务上表现良好，但其泛化能力受限于可获取且可验证的奖励数据稀缺。为解决这一问题，我们从8个维度、46个数据源构建了一个全面的强化学习就绪视觉推理数据集，涵盖信息图、数学、空间、跨图像、图形用户界面、医学、常识和通用科学等任务。提出基于影响函数的数据选择与难度过滤策略，筛选高质量样本。采用多轮强化学习与数据课程训练方法，构建了名为Vision-G1的视觉语言模型。该模型在多个视觉推理基准测试中达到领先水平，优于同规模VLM，并超越GPT-4o和Gemini-1.5 Flash等闭源模型。模型、代码与数据集已开源。
原文摘要 · Abstract (English)
Despite their success, current training pipelines for reasoning VLMs focus on a limited range of tasks, such as mathematical and logical reasoning. As a result, these models face difficulties in generalizing their reasoning capabilities to a wide range of domains, primarily due to the scarcity of readily available and verifiable reward data beyond these narrowly defined areas. Moreover, integrating data from multiple domains is challenging, as the compatibility between domain-specific datasets remains uncertain. To address these limitations, we build a comprehensive RL-ready visual reasoning dataset from 46 data sources across 8 dimensions, covering a wide range of tasks such as infographic, mathematical, spatial, cross-image, graphic user interface, medical, common sense and general science. We propose an influence function based data selection and difficulty based filtering strategy to identify high-quality training samples from this dataset. Subsequently, we train the VLM, referred to as Vision-G1, using multi-round RL with a data curriculum to iteratively improve its visual reasoning capabilities. Our model achieves state-of-the-art performance across various visual reasoning benchmarks, outperforming similar-sized VLMs and even proprietary models like GPT-4o and Gemini-1.5 Flash. The model, code and dataset are publicly available at https://github.com/yuh-zha/Vision-G1.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。