Visual language models (VLMs) rapidly progressed with the recent success of large language models. There have been growing efforts on visual instruction tuning to extend the LLM with visual inputs, but lacks an in-depth study of the visual language pre-training process, where the model learns to perform joint modeling on both modalities. In this work, we examine the design options for VLM pre-training by augmenting LLM towards VLM through step-by-step controllable comparisons. We introduce three main findings: (1) freezing LLMs during pre-training can achieve decent zero-shot performance, but lack in-context learning capability, which requires unfreezing the LLM; (2) interleaved pre-training data is beneficial whereas image-text pairs alone are not optimal; (3) re-blending text-only instruction data to image-text data during instruction fine-tuning not only remedies the degradation of text-only tasks, but also boosts VLM task accuracy. With an enhanced pre-training recipe we build VILA, a Visual Language model family that consistently outperforms the state-of-the-art models, e.g., LLaVA-1.5, across main benchmarks without bells and whistles. Multi-modal pre-training also helps unveil appealing properties of VILA, including multi-image reasoning, enhanced in-context learning, and better world knowledge.

翻译：视觉语言模型（VLM）随着大语言模型的最新成功而迅速发展。尽管在视觉指令微调方面已有大量研究以扩展大语言模型的视觉输入能力，但对视觉语言预训练过程（即模型学习对两种模态进行联合建模）的深入探讨仍显不足。本文通过逐步可控的比较实验，研究如何将大语言模型扩展为视觉语言模型时预训练的设计选择。我们提出三项主要发现：（1）预训练期间冻结大语言模型可获得不错的零样本性能，但缺乏上下文学习能力，这需要解冻大语言模型；（2）交错式预训练数据具有优势，而仅使用图像-文本对并非最优选择；（3）在指令微调阶段将纯文本指令数据与图像-文本数据重新混合，不仅能弥补纯文本任务性能的下降，还能提升视觉语言模型的任务准确率。基于优化的预训练方法，我们构建了VILA视觉语言模型系列，该模型无需复杂技巧即可在主流基准测试中持续超越当前最先进的模型（例如LLaVA-1.5）。多模态预训练还揭示了VILA的若干优异特性，包括多图像推理、增强的上下文学习能力以及更优的世界知识。