用增强版ImageNet训练出媲美海量数据模型的文生图系统
How far can we go with ImageNet for Text-to-Image generation?
- 仅用ImageNet加文本图像增强，替代超大规模网络爬取数据
- 性能超越SD-XL，参数量仅为1/10，训练图片仅1/1000
- 适合追求可复现、低资源训练的研究者与开发者
近期文生图模型通过在十亿级数据集上训练取得了显著进展，遵循‘数据越多越好’的范式，优先考虑数据量而非可用性（闭源与开源）和可复现性（数据衰减与稳定数据集）。本文挑战这一范式，证明仅使用增强版ImageNet配合精心设计的文本与图像增强，即可达到在大规模网络爬取数据上训练模型的能力。在此更简洁的设置下，我们在GenEval上得分高出SD-XL 6%，在DPGBench上高出5%，同时参数量仅为1/10，训练图像仅1/1000。我们还表明，基于ImageNet预训练的模型可在特定任务数据集（如高分辨率美学应用）上微调并取得良好效果，说明ImageNet足以获取通用视觉能力。这为更可复现的研究开辟了道路，因为ImageNet广泛可用，且所提出的标准化训练方案仅需500小时H100即可完成训练。
原文摘要 · Abstract (English)
Recent text-to-image (T2I) generation models have achieved remarkable sucess by training on billion-scale datasets, following a `bigger is better' paradigm that prioritizes data quantity over availability (closed vs open source) and reproducibility (data decay vs established collections). We challenge this established paradigm by demonstrating that one can achieve capabilities of models trained on massive web-scraped collections, using only ImageNet enhanced with well-designed text and image augmentations. With this much simpler setup, we achieve a +6% overall score over SD-XL on GenEval and +5% on DPGBench while using just 1/10th the parameters and 1/1000th the training images. We also show that ImageNet pretrained models can be finetuned on task specific datasets (like for high resolution aesthetic applications) with good results, indicating that ImageNet is sufficient for acquiring general capabilities. This opens the way for more reproducible research as ImageNet is widely available and the proposed standardized training setup only requires 500 hours of H100 to train a text-to-image model.
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。