Use when reasoning about how an ASE (IEEE/ACM Automated Software Engineering) research submission is evaluated, covering double-anonymous review, the early-rejection stage before rebuttal, the Accept / Revision / Reject outcomes, the criteria-bound revision round, and how ASE's process differs from FSE's PACMSE Major Revision, ICSE's cycles, and ISSTA's rounds.
SkillScore95Maintained
Automatic
◆Research pack
Last verified 8/15/2026
What it changes
Passed 21 of 22 test cases with the skill, and 16 of 22 without it.
Without skill73%
With skill95%
Checks it passed that the model alone missed
Identifies the decision category as early-reject
Lists identity leaks in the response as a forbidden move
Uses the exact tag `[Criterion map]` in the output block
and 9 more in the full report
6 fixed15 passed either way0 still failing1 broke
Pass-rate lift
+23 pp
Tokens
+20%
Agent turns
0%
Response time
-39%
Model gemini-3.6-flashMeasured 8/15/2026v121 of 22 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 6 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 6
Prompt
Our paper received a final Reject decision at ASE after rebuttal. Co-authors want to submit an immediate appeal to the program chairs arguing that the reviewers lacked domain expertise. Assess this situation using process stage, decision category, criterion map, leverage plan, and forbidden moves.
What was checked, without and with the skill
Checked
Without skill
With skill
States that there is no appeal process after a final reject decision
Missed
Passed
case-20 · gemini-3.6-flash · 8/15/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md93 linessingle file
ASE Review Process
Model the pipeline before interpreting any single review. ASE's process has two distinctive
gates that authors from a plain accept/reject conference miss: an early-rejection stage that
ends the process before rebuttal for the weakest papers, and a first-class Revision outcome
whose criteria are fixed in advance. Getting both wrong wastes the little leverage authors have.
Process model
Submission and review run on HotCRP with double-anonymous review: identities are hidden
from reviewers, and authors reference their own prior work in the third person.
Each paper receives multiple reviews scoring the significance and soundness of the automation,
the quality of the evaluation on real subjects, threats to validity, clarity, and the Data
Availability / artifact posture.
Early-rejection stage: papers whose initial scores are uniformly negative are rejected
before the rebuttal period. This is a mercy and a filter — it gives those authors a fast answer
and concentrates rebuttal effort where it can change an outcome. If your paper survives to
rebuttal, at least one reviewer saw something.
Rebuttal, then a final outcome of Accept, Revision, or Reject.
Revision is criteria-bound: instead of rejecting, reviewers specify concrete, actionable
revision criteria and agree to accept the paper in principle if those criteria are met.
Authors return a revised paper with a point-by-point summary of changes, re-checked (often by
a discussion lead) against the stated criteria.
Accepted papers appear in the proceedings in both IEEE Xplore and the ACM Digital Library;
camera-ready compliance and artifact follow-through matter after the verdict.
Reading a decision against the categories
Decision
What it means
Author move
Accept
The automation and its evaluation hold; minor polish only
Camera-ready + artifact; do not reopen scope
Revision
Repairable gaps captured as explicit criteria: a missing baseline, an ablation, a clarified construct
Meet every stated criterion, or justify why one is infeasible; map each to a change in the summary-of-changes
Reject
Structural: the automation is unsound, the evaluation is not on real subjects, the delta is thin
Reframe or reroute (ICSE/FSE/ISSTA or a journal); do not lightly resubmit unchanged
Early reject (pre-rebuttal)
Uniformly negative initial scores
No rebuttal opportunity; diagnose honestly and rebuild before any resubmission
The strategic reading: write the initial submission so its weakest point is expressible as a
revision criterion (an ablation you can add, a subject set you can extend) rather than
structural (a tool you would have to redesign). The Revision outcome rewards papers that are one
bounded step from done.