Recent advancements in Large Vision-Language Models (LVLMs) have enabled
processing of multimodal inputs in language models but require significant
computational resources for deployment, especially in edge devices. This study
aims to bridge the performance gap between traditional-scale LVLMs and
resource-friendly lite versions by adopting high-quality training data. To do
this, a synthetic dataset is created by leveraging GPT-4V's ability to generate
detailed captions, complex reasoning instructions and detailed answers from
images. The resulted model trained with our data, ALLaVA, achieves competitive
performance on 12 benchmarks up to 3B LVLMs. This work highlights the
feasibility of adopting high-quality data in crafting more efficient LVLMs. Our
online demo is available at https://allava.freedomai.cn.

利用 GPT-4V 生成图像的详细标题、复杂的推理指令和详细答案，通过合成数据集，我们训练了 ALLaVA 模型，该模型在 12 个基准测试中取得了竞争性的性能，展示了在构建更高效的 LVLMs 中采用高质量数据的可行性。