ECAI Workflow
Plan the ECAI year backward from the paper deadline, because ECAI is a single-round process
with a two-deadline front (abstract, then paper) and no revision round to catch a gap later.
First fix which regime the edition is in — it changes the calendar owner, the template, and whether
PAIS is available:
- Standalone ECAI (e.g. 2024): EurAI + a host society; FAIA/
ecai.cls; PAIS co-located.
- IJCAI-ECAI 2026: joint with IJCAI in Bremen (15-21 Aug 2026); IJCAI process/template/proceedings.
The IJCAI-ECAI 2026 calendar (verify on the live page)
Abstract deadline .............. 12 January 2026 (AoE) <- title/abstract/authors/topics locked
Full paper deadline ............ 19 January 2026 (AoE) <- anonymized PDF + supplement
Summary-reject notification .... 4 March 2026 <- early filter; below-bar/out-of-scope out
Author response (rebuttal) ..... 7-10 April 2026 <- one short window, factual corrections
Final notification ............. 29 April 2026 <- accept / reject (no revision round)
Camera-ready ................... (per author kit) <- de-anonymize, open-access production
Conference ..................... 15-21 August 2026, Bremen
For a standalone ECAI the shape is the same (registration ~1 week before the PDF; double-blind
review with a rebuttal; open-access FAIA proceedings) but the dates and organizer differ — read the
current call.
Backward plan from the paper deadline
- Deadline − 12 weeks: decide ECAI vs IJCAI/AAAI and main track vs PAIS
(
ecai-topic-selection). Routing before writing is the highest-leverage move, because the 7-page
budget and the venue's breadth shape the paper.
- − 10 weeks: lock the contribution and the evidence type (proof vs experiment) per claim
shape (
ecai-experiments); name threats/assumptions now — you cannot add them in a later round.
- − 8 weeks: draft the body to the first-page arc and the 7-page budget
(
ecai-writing-style); start the anonymized supplement (ecai-reproducibility).
- − 4 weeks: finish full proofs / complete experiments with seeds and spread; write delta-first
related work (
ecai-related-work).
- − 2 weeks: internal review; run the double-blind sweep and the body/supplement split
(
ecai-supplementary).
- Abstract deadline: register title/abstract/authors/topics with the real text.
- Paper deadline: run
ecai-submission end to end on the final PDF, supplement, template, and
per-author cap.
After submission
- Summary-reject phase: nothing to do but ensure you did not invite it — scope, formatting,
anonymity, and page budget must have been correct at upload.
- Rebuttal window:
ecai-author-response — one short, precise, double-blind reply; fix factual
misreadings and supply asked-for numbers; promise only minor camera-ready edits.
- Notification:
- Accept →
ecai-camera-ready (de-anonymize, correct template/publisher regime, open-access
forms, DOI/citation, register to present) and make the supplement a permanent open release
(ecai-artifact-evaluation).
- Reject → revise substantively and reroute (AAAI/IJCAI/specialist venue) or aim at the next
ECAI; there is no appeal that reopens the decision (
ecai-review-process).
Parallel tracks to run early
- Reproducibility is not a last-minute task. Seeds, cached outputs, and full proofs must be
produced as you do the work; they cannot be retrofitted at the deadline (
ecai-reproducibility).
- PAIS decision is a fork, not an afterthought. A deployed-application paper aimed at PAIS
leads with real-world impact and has its own call — decide before the abstract deadline.
- Per-author submission cap (IJCAI-ECAI 2026: 8) — coordinate across your group early so you do
not have to withdraw a paper at the deadline.
Milestone checklist
[ ] Regime fixed: standalone ECAI (FAIA/PAIS) or joint IJCAI-ECAI (IJCAI)
[ ] Track chosen: main track vs PAIS
[ ] Evidence type set per claim (proof vs experiment); threats/assumptions named
[ ] Body drafted to 7-page arc; supplement started and anonymized
[ ] Abstract registered by the earlier deadline
[ ] Double-blind sweep + body/supplement split done before paper upload
[ ] Rebuttal plan ready for the response window
[ ] Camera-ready path (template/publisher/forms/DOI/registration) mapped
Reverify each cycle
Regime (standalone vs joint), all dates and AoE cutoffs, the per-author cap, the ethics/AI policy,
and the camera-ready production instructions — none are stable across editions.
1---2name: ecai-workflow3description: Use to plan an ECAI campaign end to end — run the single-round year backward from the paper deadline through the abstract/registration step, summary-reject phase, rebuttal window, notification, and the open-access camera-ready, adapting to whether the edition is a standalone ECAI (FAIA/PAIS) or the joint IJCAI-ECAI 2026.4---56# ECAI Workflow78Plan the ECAI year **backward from the paper deadline**, because ECAI is a **single-round** process9with a **two-deadline** front (abstract, then paper) and **no revision round** to catch a gap later.10First fix which regime the edition is in — it changes the calendar owner, the template, and whether11PAIS is available:1213- **Standalone ECAI** (e.g. 2024): EurAI + a host society; FAIA/`ecai.cls`; PAIS co-located.14- **IJCAI-ECAI 2026**: joint with IJCAI in Bremen (15-21 Aug 2026); IJCAI process/template/proceedings.1516## The IJCAI-ECAI 2026 calendar (verify on the live page)1718```text19Abstract deadline .............. 12 January 2026 (AoE) <- title/abstract/authors/topics locked20Full paper deadline ............ 19 January 2026 (AoE) <- anonymized PDF + supplement21Summary-reject notification .... 4 March 2026 <- early filter; below-bar/out-of-scope out22Author response (rebuttal) ..... 7-10 April 2026 <- one short window, factual corrections23Final notification ............. 29 April 2026 <- accept / reject (no revision round)24Camera-ready ................... (per author kit) <- de-anonymize, open-access production25Conference ..................... 15-21 August 2026, Bremen26```2728For a standalone ECAI the shape is the same (registration ~1 week before the PDF; double-blind29review with a rebuttal; open-access FAIA proceedings) but the dates and organizer differ — read the30current call.3132## Backward plan from the paper deadline3334- **Deadline − 12 weeks:** decide **ECAI vs IJCAI/AAAI** and **main track vs PAIS**35 (`ecai-topic-selection`). Routing before writing is the highest-leverage move, because the 7-page36 budget and the venue's breadth shape the paper.37- **− 10 weeks:** lock the contribution and the **evidence type** (proof vs experiment) per claim38 shape (`ecai-experiments`); name threats/assumptions now — you cannot add them in a later round.39- **− 8 weeks:** draft the body to the **first-page arc** and the 7-page budget40 (`ecai-writing-style`); start the anonymized supplement (`ecai-reproducibility`).41- **− 4 weeks:** finish full proofs / complete experiments with seeds and spread; write delta-first42 related work (`ecai-related-work`).43- **− 2 weeks:** internal review; run the **double-blind sweep** and the **body/supplement split**44 (`ecai-supplementary`).45- **Abstract deadline:** register title/abstract/authors/topics with the *real* text.46- **Paper deadline:** run `ecai-submission` end to end on the final PDF, supplement, template, and47 per-author cap.4849## After submission5051- **Summary-reject phase:** nothing to do but ensure you did not invite it — scope, formatting,52 anonymity, and page budget must have been correct at upload.53- **Rebuttal window:** `ecai-author-response` — one short, precise, double-blind reply; fix factual54 misreadings and supply asked-for numbers; promise only minor camera-ready edits.55- **Notification:**56 - **Accept** → `ecai-camera-ready` (de-anonymize, correct template/publisher regime, open-access57 forms, DOI/citation, register to present) and make the supplement a permanent open release58 (`ecai-artifact-evaluation`).59 - **Reject** → revise substantively and reroute (AAAI/IJCAI/specialist venue) or aim at the next60 ECAI; there is no appeal that reopens the decision (`ecai-review-process`).6162## Parallel tracks to run early6364- **Reproducibility is not a last-minute task.** Seeds, cached outputs, and full proofs must be65 produced *as you do the work*; they cannot be retrofitted at the deadline (`ecai-reproducibility`).66- **PAIS decision is a fork, not an afterthought.** A deployed-application paper aimed at PAIS67 leads with real-world impact and has its own call — decide before the abstract deadline.68- **Per-author submission cap** (IJCAI-ECAI 2026: 8) — coordinate across your group early so you do69 not have to withdraw a paper at the deadline.7071## Milestone checklist7273```text74[ ] Regime fixed: standalone ECAI (FAIA/PAIS) or joint IJCAI-ECAI (IJCAI)75[ ] Track chosen: main track vs PAIS76[ ] Evidence type set per claim (proof vs experiment); threats/assumptions named77[ ] Body drafted to 7-page arc; supplement started and anonymized78[ ] Abstract registered by the earlier deadline79[ ] Double-blind sweep + body/supplement split done before paper upload80[ ] Rebuttal plan ready for the response window81[ ] Camera-ready path (template/publisher/forms/DOI/registration) mapped82```8384## Reverify each cycle8586Regime (standalone vs joint), all dates and AoE cutoffs, the per-author cap, the ethics/AI policy,87and the camera-ready production instructions — none are stable across editions.
brycewang-stanford