Answers are grounded in this paper and cite their sources.
We fix, before the release of Gaia DR4 (2026-12-02), the complete procedure of a search for stars with single, deep and long dimming events in Gaia epoch photometry: candidate-selection tiers, a dimming detector (B2), artefact flags, cross-checks and reporting rules. Using the real Gaia DR3 transit times and the nominal DR4 scanning law we measure the recovery efficiency for injected dimmings of 1 to 1000 days. The DR3 selection based on vari_summary statistics acts as a band-pass filter for durations of about 30-300 days and almost never recovers multi-year events such as those of epsilon Aurigae; the B2 detector recovers 25% of 1000-day, 1 mag dimmings with the real DR3 cadence and about 80% with the nominal DR4 law, and flags 5% of random Gaia variables. A cross-match with the AAVSO Variable Star Index shows that 53 of our 55 best DR3 candidates are already catalogued, so no discovery is claimed. Seven numerical predictions that Gaia DR4 can confirm or falsify are registered, together with the rules for reporting the results. The deposit contains the note (PDF and Markdown), figures, the tables behind every number (CSV) and the Python code. The analysis was carried out with the assistance of an AI model (Claude, Anthropic); see "Author contributions and use of AI" in the note.
No takes yet. Share an insight, caveat, or question.
Antonio Acri (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: