实现跟跟论文图不一致

[iFe1er]

实现跟跟论文图不一致

[kaiwenzha]

Hi, thank you for your interest in our work.

To clarify, Fig. 2 is meant to illustrate our core idea, and for simplicity, we do not show the two augmented views for each sample. However, in the text descriptions of our approach (see Sec. 3 in our paper), we have clearly stated that our approach operates on a two-view batch, which is consistent with our implementation.

Regarding your concern about the effect of data augmentation, we've already discussed it in Sec. 5.4 and conducted detailed ablation studies in Appendix G.3, in which we have shown that our method consistently outperforms the baseline with or without data augmentation. This confirms that the performance gain of RnC stems from our proposed loss rather than the data augmentation.