fcbc95ccdd551da181207c0c1400c655-Supplemental.pdf

A When Do Bigger Models Help More? Figure A.1 shows relative improvement by increasing the model size under different amount of It is also worth noting that these results may reflect a "ceiling effect": as the performance gets closer Figure A.1: Relative improvement (top-1) when model size is increased. Figure B.1 shows the top-1 accuracy of fine-tuned SimCLRv2 models of different sizes. For fine-tuning on 1% of labels, SK is much more efficient. Figure C.1 shows the correlation under two different fine-tuning strategies: We observe that overall there is a linear correlation. Furthermore, as label fraction increases, the slope is decreasing.