AutOA ESWA Submission Version (30-run Benchmark)

This release corresponds to the version of the code used in the ESWA manuscript submission.

All experiments were conducted with:

• 30 independent runs per algorithm
• Fixed function evaluation (FE) budget
• Mann–Whitney U statistical testing
• Equal population size across methods

This release freezes the experimental configuration for reproducibility.