Added Author Page for Ji Wang

[weissenh]

TL;DR Disambiguating 3 papers for "Ji Wang" into a catch-all ji-wang with two papers and one author ji-wang-nudt with one paper.

(Please replace this text with a description of the changes effected by this pull request.
Include a link to the corresponding Github Issue, if there is one.
Details on how to do this (can be found here).)

Closes #3450

Interestingly, all 3 papers list as affiliation National University of Defense Technology in China.
But the 2019 papers prepend this with "HPCL, College of Computer," and have a slightly different email handle ("wj" vs "wangji"), whereas the 2021 says "Changsha, China" instead of just "China" as location. Neither paper mentions an ORCID in the XML for this name.

• https://aclanthology.org/2021.findings-acl.156.pdf (NUDT) "Gaussian Process..."
• https://aclanthology.org/P19-1440.pdf (catch-all) "Complex Question Decomposition ..."
• https://aclanthology.org/K19-1074.pdf (catch-all) "Pretraining-based Natural Language ..."

The issue submitter hasn't mentioned their ORCID or degree institution or provided any links, nor is their GitHub page informative. The issue submitter said

The following two papers belong to other people.
Complex Question Decomposition for Semantic Parsing
Pretraining-Based Natural Language Generation for Text Summarization

On https://openreview.net/profile?id=~Ji_Wang3 I see the catch-all ACL Anthology linked, NUDT affiliation throughout career, Google Scholar and DBLP link. GS and DBLP show the ACL-IJCNLP 2021 paper ("Gaussian Process ..."). The DBLP page further references an ORCID, but the orcidg.org page doesn't list the paper.

3 papers: https://aclanthology.org/people/ji-wang/
After changes:

• https://preview.aclanthology.org/author-page-ji-wang/people/ji-wang/
• https://preview.aclanthology.org/author-page-ji-wang/people/ji-wang-nudt/

[mjpost]

Typically we do not process these requests unless the user submits an ORCID iD and affiliation as described. I think it's good to do if only because it saves a lot of work which may also introduce mistakes.

[weissenh]

Typically we do not process these requests unless the user submits an ORCID iD and affiliation as described. I think it's good to do if only because it saves a lot of work which may also introduce mistakes.

I understand this point. And I've observed that it can take time to research this information myself - unless this information is easily obtained (XML contains ORCID, orcid.org page contains detailed career history and further links to verify the information).
For the future (not counting my existing PRs) I'll stop doing this or ask issue submitters for this information first.

@mjpost What should we do with this PR then? Should I convert it to draft mode and ask the issue submitter for affiliation and ORCID first? (Not sure when they will respond - the issue has been opened nearly one and a half years ago)