Dear Authors,
Thank you for sharing the code accompanying your brilliant work! I deeply appreciate your efforts to make your research more accessible.
However, after numerous attempts, I have been unable to successfully run the training scripts. Specifically, I encountered issues with the overlap loss implementation. Even after removing the overlap loss described in your paper, the code still does not function as expected.
Could you kindly review the released code to address these issues? Your guidance or corrections would be greatly appreciated.
Thank you for your time and assistance.
Dear Authors,
Thank you for sharing the code accompanying your brilliant work! I deeply appreciate your efforts to make your research more accessible.
However, after numerous attempts, I have been unable to successfully run the training scripts. Specifically, I encountered issues with the overlap loss implementation. Even after removing the overlap loss described in your paper, the code still does not function as expected.
Could you kindly review the released code to address these issues? Your guidance or corrections would be greatly appreciated.
Thank you for your time and assistance.