pcmepp icon indicating copy to clipboard operation
pcmepp copied to clipboard

Official Pytorch implementation of "Improved Probabilistic Image-Text Representations" (ICLR 2024)

Results 2 pcmepp issues
Sort by recently updated
recently updated
newest added

![image](https://github.com/naver-ai/pcmepp/assets/55874332/24a271d0-334e-4e75-b900-2d15effc071b) Hello! While studying your code, I found that when resampling is used here, it seems to be calculated directly using the variance. In my understanding, resampling seems to be...

![image](https://github.com/naver-ai/pcmepp/assets/55874332/b434439a-7df8-4c45-89f9-4419accbe515) For binary loss, the final derivative results seem to be different from the paper, when m=0 or 1, I calculated the results as -sigmoid(l_vt) and sigmoid(-l_vt), respectively.