Scaling laws have been recently employed to derive compute-optimal model size (number of parameters) for a given compute duration. We advance and refine such methods to infer compute-optimal model shapes, such as width and depth, and successfully implement this in vision transformers. Our shape-optimized vision transformer, SoViT, achieves results competitive with models that exceed twice its size, despite being pre-trained with an equivalent amount of compute. For example, SoViT-400m/14 achieves 90.3% fine-tuning accuracy on ILSRCV2012, surpassing the much larger ViT-g/14 and approaching ViT-G/14 under identical settings, with also less than half the inference cost. We conduct a thorough evaluation across multiple tasks, such as image classification, captioning, VQA and zero-shot transfer, demonstrating the effectiveness of our model across a broad range of domains and identifying limitations. Overall, our findings challenge the prevailing approach of blindly scaling up vision models and pave a path for a more informed scaling.

翻译：缩放定律近期被用于在给定计算时长下推导计算最优的模型大小（参数量）。我们推进并完善了此类方法，以推断计算最优的模型形状（如宽度和深度），并成功将其应用于视觉Transformer中。我们提出的形状优化视觉Transformer——SoViT，虽然预训练时使用的计算量相同，但其性能可与参数量两倍于自身的模型相媲美。例如，SoViT-400m/14在ILSRCV2012上的微调准确率达到90.3%，超过了更大的ViT-g/14，并在相同设置下接近ViT-G/14，而推理成本却不到后者的一半。我们在图像分类、图像描述、VQA和零样本迁移等多个任务上进行了全面评估，证明了模型在广泛领域中的有效性，并指出了其局限性。总体而言，我们的研究结果挑战了盲目扩大视觉模型的普遍做法，并为更科学的缩放策略铺平了道路。