Changelog

What changed in the datasets, baselines, and published results — recorded per edition, so you can see exactly what differs from year to year.

Notable changes to the OAEI Bio-ML datasets, baselines, and published results are recorded here, per edition.

2026 edition (first edition)

2026-07-07 — Finalised datasets published

The 2026 datasets are frozen and distributed on the Hugging Face Hub as OAEI-ML/bio-ml (edition tag 2026): the per-pair equivalence reference (refs_equiv/train.tsv), the local- and typed-ranking candidate pools (with gold-stripped test splits), and a repaired/ tree against the coherence-repaired reference. The Hugging Face dataset is data only; the self-contained scoring_kit/ ships separately with the track repository. Three CodaBench competitions (Track 1 Global Alignment, Track 1 Local Ranking, Track 2 Typed Ranking) open on 12 July 2026; the evaluation window runs 12 July – 1 September 2026.

2026-07-06 — Provisional materials released

Provisional website, task documentation, and datasets made public for review, together with the organiser-run baselines on the baselines page.

Entry format for future editions

From the 2027 edition onward, entries will record, relative to the previous edition:

  • Dataset changes — ontology version bumps, reference additions/removals/repairs (with correspondence counts), candidate-pool regeneration, and new dataset revisions.
  • Protocol changes — metric, validation, or submission-format changes.
  • Results — a pointer to the closed edition’s final archived results.