Confusion on papers & evaluation

[zp9187]

Hi Wang,
I'm interested in your awesome work of VIES, but i have the following confusion and am looking forward to your reply.

1. In the sequence labeling section, the input is features sequence of each text segment or flattened texts of whole image?
2. In the evaluation, you evaluate only the entities of value or entities of both key and value.
3. Do you have the plan to release the source code?

Many thanks.

[zp9187]

@jpWang