Agent Skills
› brycewang-stanford/Awesome-Journal-Skills
› edbt-submission
edbt-submission
GitHub用于在上传至Microsoft CMT前审计EDBT研究类投稿，涵盖周期选择、论文类型（Regular/Experiments & Analysis/Vision）、OpenProceedings模板与页数限制、12个月重投禁令及可复现性检查。
Trigger Scenarios
准备向 Microsoft CMT 提交 EDBT 论文
需要确认 EDBT 投稿格式与政策合规性
Install
npx skills add brycewang-stanford/Awesome-Journal-Skills --skill edbt-submission -g -y
SKILL.md
Frontmatter
{
"name": "edbt-submission",
"description": "Use when auditing an EDBT research-track submission for Microsoft CMT readiness, covering the cycle choice, the paper shape (Regular \/ Experiments-&-Analysis \/ Vision), the OpenProceedings\/host LaTeX template and per-shape page budget, the 12-month resubmission ban, reproducibility items, and desk-reject triage before the cycle deadline."
}
EDBT Submission
Run this audit before uploading to Microsoft CMT for an EDBT cycle. EDBT is the European
database-systems flagship; its papers are published open access on OpenProceedings.org under a
CC-BY-NC-ND license, and it runs a multiple-cycle rolling model rather than one annual
deadline. Every number below was read from the EDBT/ICDT 2026 (Tampere) call and the EDBT 2022
research CFP on 2026-07-09 via search renderings (see resources/official-source-map.md); treat
them as a one-cycle snapshot and reopen the live call first.
Pick the cycle and the paper shape first
Two decisions precede the audit, and both change the checklist:
- Which cycle? EDBT runs three submission/publication cycles a year (since 2022). A paper accepted in an earlier cycle presents at that year's conference; the last cycle rolls to the next edition. Choose the cycle your evidence is actually ready for — a rushed submission into an early cycle wastes a revise opportunity you could have earned with two more weeks of measurement. (EDBT 2027 cycle deadlines: 4 Feb 2026, 10 Jun 2026, 7 Oct 2026; cycle 3 is the live one as of 2026-07-09.)
- Which shape? EDBT distinguishes Regular research papers, Experiments & Analysis papers (where the benchmarking/repeatability study is the contribution), and short Vision papers. The shape sets the page budget and what reviewers weigh — do not submit an Experiments & Analysis paper as a Regular one or vice versa.
Format and page budget
- Host/OpenProceedings LaTeX template, unmodified. EDBT proceedings are produced by
OpenProceedings, not ACM DL or IEEE Xplore — use the template the current host site ships, not an
acmartorIEEEtranfile carried over from a US venue. - Page budget (verify per cycle): from the EDBT 2022 research CFP, Regular and Experiments & Analysis papers are ≤12 pages plus unlimited references; Vision papers are ≤6 pages plus unlimited references. References do not count against the limit, but figures, tables, and appendices inside the body do. (待核实 for the exact current-cycle numbers.)
- Margins, font size, and spacing are fixed by the template; editorial compression, not template tampering, is how you recover space.
The resubmission-ban check (EDBT-specific)
EDBT enforces a 12-month resubmission ban: work rejected or withdrawn from an EDBT track cannot be resubmitted to any EDBT track in the same paper format for 12 months. Before you submit:
- Confirm this exact work has not been rejected/withdrawn from an EDBT track in the past year in the same format. A Regular paper reworked as a Vision paper is a different format — but do not guess the boundary; check the current wording.
- This is not a dual-submission rule (that is separate); it is EDBT's own re-entry cooldown, a direct consequence of the rolling-cycle model, where nothing stops rapid re-tries otherwise.
Blind policy and identity
- Confirm on the current call whether reviewing is single-blind or double-blind (待核实; EDBT has historically used author-identified single-blind reviewing). If the cycle is double-blind, anonymize the PDF, the artifact, and any repository link:
# If the current cycle is double-blind, run before upload:
pdfinfo paper.pdf | grep -Ei 'author|creator|producer'
pdftotext paper.pdf - | grep -nEi 'github\.com/[a-z0-9-]+|zenodo\.org/record|acknowledg|grant' | head
unzip -l artifact.zip | grep -Ei '\.git/|\.DS_Store|/home/|/Users/' | head
If the cycle is single-blind, skip anonymization but still scrub broken author metadata and stale paths from the artifact.
Reproducibility at submission time
EDBT's community values inspectable, reproducible work, and the published record is open access:
- Provide a runnable artifact or a clear availability statement: what exists, the workloads and dataset versions, the hardware/cluster configuration, and where the package will live.
- For an Experiments & Analysis paper this is not optional decoration — the repeatability of the measurements is the contribution, so the package and the paper must line up exactly.
- Pin dataset versions, query/workload provenance, and engine build/commit now; these cannot be reconstructed after the cycle closes.
Desk-risk triage
| Finding at audit time | Severity | Real fix |
|---|---|---|
| Body over the per-shape page budget | Desk-reject-grade | Cut or move to the artifact; references do not absorb body text |
Wrong template (acmart/IEEEtran instead of the host/OpenProceedings style) |
Format violation | Recompile in the correct template |
| Paper shape mismatched to contribution | Mis-routed review | Re-declare Regular / Experiments-&-Analysis / Vision correctly |
| Work under the 12-month EDBT ban in this format | Ineligible | Do not submit; wait out the cooldown or change format only if the rules allow |
| Double-blind cycle with identity leak | Anonymity violation | Re-anonymize PDF and artifact |
| No artifact / availability story | Scored against the paper (esp. Experiments & Analysis) | Add a runnable package or an explicit, justified exception |
| Same work under review elsewhere | Dual-submission exposure | Withdraw one venue; verify the concurrent-submission wording |
Final-week order of operations
- Freeze the body early; the argument and the measurements cannot churn late.
- Confirm the cycle, the paper shape, and the resubmission-ban status against the live call.
- Build the artifact and pin every workload, dataset version, and hardware detail.
- Compile clean in the host/OpenProceedings template; check the per-shape page budget.
- Fill every CMT field — subject areas that match your evidence, conflicts for every coauthor's institution and recent collaborators — a day early; late conflicts are the classic midnight failure.
- Re-download the uploaded PDF and read it cold to confirm it is the file you meant.
Reverify each cycle
- The number of cycles and the exact per-cycle deadline, author-feedback, and revision dates.
- The page budget and template revision per paper shape, and whether a short-paper category exists.
- Single- vs double-blind, the reproducibility/availability requirement, and any AI-disclosure rule — all cycle-volatile; none should be assumed stable.
Output format
[EDBT submission status] ready / blocked / needs work
[Cycle + shape] cycle N (<deadline>) / Regular | Experiments-&-Analysis | Vision
[Eligibility] resubmission-ban clear? yes/no
[Format] pages used (body), template compliance, per-shape budget met?
[Blind] policy this cycle + (if double-blind) anonymity clean/leaks
[Reproducibility] runnable artifact? provenance pinned?
[Fix queue] <ordered, with owners and dates before the cycle deadline>
Version History
- 9f86f09 Current 2026-07-19 15:16
