Important Notes:

• Teams that did not submit a paper have been excluded from the official ranking, as their methodologies could not be verified.

• For each team, we considered only their two most recent submissions per task (per dataset, in the case of Task C). Among those two, we selected the single best-performing submission to represent the team on the leaderboard — ranked by Averaged Graph Similarity for Flagship and Reuse tasks, and by Averaged F1 Score for Taxonomy Task. This means the leaderboard shows one row per team: their strongest recent result, not just their latest one. Teams with only a single submission had that one submission evaluated directly. 

Flagship Task Leaderboard (FINAL)

Reuse Task Leaderboard (FINAL)

Taxonomy Learning Task Leaderboard (FINAL)