Zhuang Li

Hi, I looked into the code of RATSQL. It seems the initial state of the decoder is zero instead of the end state of the questions. Am I understanding this...

Hi Sometimes, the solution I got does not fit the constraints I set. But when I set an initial solution for the problem, the problem was solved correctly. Why is...

customized dataset

1

Hi sorry for this naive question as I am new to computer vision and DALLE. Say if I have a folder of images and captions such as dir/ cat.jpg cat.txt...

In the evaluation, I found that you only selected 10 candidate labels for evaluation. I this this evaluation method is quite odd. Can you explain why you chose this evaluation...

Several test cases the sempre seems to not be able to resolve.

1

Hi, I am conducting research based on the [semantic parsing paper of EMNLP 2013](http://www.aclweb.org/anthology/D13-1160). However, I found that there are several cases that sempre can't solve. How the sempre solve...

Hi This is work is really intereting. May you provide a well-trained model for this so I can directly replicate your experiment results? Particularly the pre-trained auto-encoder. Best Zhuang Li

Hi, I saw you implemented the supervised attention while this mechanism was not described in the paper. Is supervised attention used in your experiments and how it is performed? Thank...

Hi, I am investigating this code and want to use it as my baseline. The method seems to be neat. But I wonder why there is a re-train step in...

As far as I know, the WizardCoder is trained on GPT3.5-generated data. But why is it using a llama or OpenRAIL-M license? Anything that is tuned using ChatGPT-generated data will...

Hi I am trying to run your code but am having difficulty replicating the result in IJCAI 2018. Why is the valid set the same with the test set in...