Offline Multi-Objective Optimization on Off-MOO-Bench Sci-Design

1.8Average IGDoffline Rank

D(best)

Updated 1mo ago

Evaluation Results

Method	Links
D(best) 2026.06		1.8
ParetoFlow 2026.06		3.21
MOBO 2026.06		4.95
Multiple Models 2026.06		6.1
Multi Head + GradNorm 2026.06		6.2
ParetoFlow 2026.06		6.21
Multi Head 2026.06		6.25
Multiple Models 2026.06		6.45
MultipleModels + COMs 2026.06		6.72
End-to-End 2026.06		6.97
Multi Head 2026.06		7.17
Multiple Models 2026.06		7.4
End-to-End + PcGrad 2026.06		7.47
DOMOO 2026.06		7.55
Multiple Models 2026.06		7.97
MOBO-qParEGO 2026.06		8
Multiple Models + RoMA 2026.06		8.15
MOBO 2026.06		8.3
Multi Head + PcGrad 2026.06		8.35
DOMOO 2026.06		8.6
End-to-End 2026.06		8.68
Multi Head 2026.06		8.95
End-to-End 2026.06		8.97
End-to-End 2026.06		9.25
Multi Head 2026.06		9.25
Multiple Models + ICT 2026.06		9.55
End-to-End + GradNorm 2026.06		9.57
Multiple Models 2026.06		9.57
D(best) 2026.06		9.62
Multiple Models + IOM 2026.06		9.78
MOBO 2026.06		10.27
Multiple Models + Tri-Mentoring 2026.06		10.3
Multiple Models 2026.06		11.03
Multiple Models 2026.06		11.62