Turn an accepted WSDM paper into its archival form in the ACM Digital Library. WSDM proceedings are ACM proceedings (WSDM '26 is DOI 10.1145/3773966; each edition gets its own volume), so the mechanics run through ACM's standard production chain. Warning up front: the pack could not verify the current edition's camera-ready deadline or final page allowance - both are 待核实 items that the acceptance email and author-kit page settle, and nothing here substitutes for them.
1. Read the acceptance email twice; extract deadline, page allowance,
required class options, and the rights-form sender to whitelist.
2. Complete the ACM e-rights form FIRST - it generates the exact
copyright block and \setcopyright values your LaTeX must carry.
3. De-anonymize: author block, third-person self-citations back to
first person where natural, acknowledgements and grants restored.
4. Restore real artifact URLs (the anonymous repo becomes the public
one) and re-run every link.
5. Add/verify CCS concepts and keywords via the ACM CCS tool.
6. Rebuild with the camera-ready class options; check the allowance.
7. Submit source (TAPS-style production expects LaTeX/Word source,
not just a PDF); fix validation errors early - TAPS queues clog
near deadlines.
8. Register an author and confirm who presents (talk and/or poster).
Steps 2 before 3-6 matters: the rights confirmation dictates copyright text, DOI string, and reference format, and TAPS-produced output will be rejected or regenerated if the rights block is wrong.
Removing anonymous from the class options is the trivial part. The real work:
wsdm-submission)
can now be made precise if your organization's disclosure review permits it.
Industry authors: start internal publication approval the day of acceptance -
it is the most common cause of missed camera-ready deadlines, not LaTeX.Because WSDM historically has no rebuttal, there is no author-committed change list from a discussion phase. The reviews and the SPC meta-review are still a soft contract: implement the factual and presentational fixes they request, and keep a one-page internal log of what changed. If the meta-review conditions acceptance on specific fixes, treat those as hard requirements and make them findable (a sentence in the right section beats a silent table tweak).
| Item | Check | Common failure |
|---|---|---|
| Rights form | Completed by the listed corresponding author | Form sat in a spam folder |
| Copyright block | Matches e-rights output exactly | Stale \setcopyright from a template |
| Page allowance | Camera-ready cap per the author kit (待核实 each edition) | Assuming the submission cap carries over |
| CCS + keywords | Generated from the CCS tool, pasted as XML/commands | Hand-typed concepts that fail validation |
| Metadata | Title/authors in the form == PDF exactly, including accents | Unicode mismatch breaking DL indexing |
| Artifact links | Public repo resolvable, tagged release, license file | Link still pointing at the anonymous mirror |
| Ethics section | Retained in final form | Deleted to reclaim space |
| ORCID | Current ACM production requests ORCIDs for authors | One coauthor without an ORCID at deadline |
\ref in the produced PDF, not your local build.Day 0-1 acceptance email parsed; e-rights kicked off; disclosure
review opened (industry authors)
Day 2-5 de-anonymization edit pass; meta-review fixes implemented
Day 6-8 artifact goes public: real repo, license, tagged release,
archival snapshot; links restored in the source
Day 9-11 CCS concepts, ORCID sweep, metadata parity check
Day 12 production submission; validation errors fixed same day
Day 13-14 buffer for the error you did not predict
[Camera-ready status] on track / at risk (deadline: <from acceptance email>)
[Rights] e-rights done -> copyright block synced: yes / no
[De-anonymization] author block / self-citations / acks / links: done or pending
[Allowance] final page count vs verified cap: pass / over / cap unverified
[Meta-review conditions] <fix -> location implemented>
[Presence] registrant, presenter, visa lead time, disclosure approval status