---

name: vldb-submission description: Use when auditing a PVLDB submission before a monthly deadline, covering the mandatory abstract on the 25th, the CMT window, the 1st-of-month 5:00 PM Pacific cutoff, category and page-budget choice, single-blind cover-page requirements, the per-month and per-year author caps, and desk-reject hazards specific to VLDB.

VLDB Submission

Run this in the two weeks before a PVLDB monthly deadline. VLDB research papers enter through PVLDB (Proceedings of the VLDB Endowment), and the volume's own guidelines page — not the conference site's summary — is the governing document. Everything dated below was checked 2026-07-08; reopen the live volume page first.

The three clocks of one deadline

PVLDB splits a single submission across three dates, and missing the first one kills the cycle quietly:

1. 20th of the prior month — CMT opens for the cycle.
2. 25th of the prior month — mandatory abstract (title, authors, abstract, subject areas) registered in CMT. No abstract, no paper slot.
3. 1st of the month, 5:00 PM Pacific Time — full paper due. This is a Pacific-clock deadline, not AoE; for most of the world it lands in the middle of the night before the date your calendar shows. No extensions.

Category and budget check

Pick the category deliberately — reviewers calibrate expectations to it:

Category (VLDB 2026 cycle)	Body budget	What reviewers expect
Regular Research	12 pp + refs	New technique with systems evidence
Experiment, Analysis & Benchmark	12 pp + refs	Rigorous study; reproducibility evaluation is mandatory
Scalable Data Science	8 pp + refs	Data-science pipeline at scale, practice-forward
Vision	6 pp + refs	Agenda-setting argument, less evidence

References are uncapped, but appendices and acknowledgements count inside the budget. Category continuation and budgets for the volume you target: confirm on the live page (待核实 for Vol 20 beyond the sources in the source map).

Identity and caps

• PVLDB review is single-blind: author names and affiliations belong on the paper. Do not strip them the way SIGMOD or ML venues require — an anonymous PVLDB submission is a formatting error, not a virtue. Reconfirm the wording on the live guidelines before submission.
• Count coauthor load before promising a submission: Vol 20 caps each author at two papers per month and twelve per year.
• Declare conflicts accurately in CMT; the review board assignment runs on them.

Pre-deadline audit sequence

[ ] Live volume identified (Vol 20 on 2026-07-08) and its guidelines reopened
[ ] Abstract registered in CMT before the 25th, title/author list final
[ ] Category chosen; page budget measured with appendix included
[ ] Volume template applied unmodified; fonts legible at print size
[ ] Author names and affiliations present (single-blind), conflicts declared
[ ] Caps checked for every author, not just the lead
[ ] Artifact URL decided: include now or forfeit the availability signal
[ ] Deadline converted to your local time from 5:00 PM PT on the 1st

What draws a desk rejection here

• Overrunning the category budget or squeezing with template edits.
• Submitting to a closed cycle because the abstract step was missed.
• A cover page missing authors, or CMT metadata contradicting the PDF.
• Blowing an author's cap — the excess paper, not the earliest, is at risk.

One month's miss costs exactly one month — the next cycle opens on the 20th. Submitting a half-ready paper to "make this month" wastes a cap slot and burns the one revision the process will ever offer.

Output format

[Cycle] <volume / month targeted>
[Abstract stage] registered / at risk / missed
[Category + budget] <category, pages used vs. limit>
[Identity check] single-blind cover page complete / issues
[Cap status] <per-author counts vs. month and year caps>
[Blocking fixes] <ordered, with owner>