# Offline Multi-Objective Optimization on Off-MOO-Bench MO-NAS (50th percentile)

Source: [Offline Multi-Objective Optimization on Off-MOO-Bench MO-NAS (50th percentile)](<https://www.sota2.com/research/sota/offline-multi-objective-optimization-on-off-moo-bench-mo-nas-50th-percentile>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-06-16T07:29:18.693Z

Evaluation of Offline Multi-Objective Optimization algorithms specifically on the 50th percentile results of the MO-NAS component of the Off-MOO-Bench suite, measured using the Average HV rank metric.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Offline Multi-Objective Optimization |
| Dataset (including any stated split) | Off-MOO-Bench MO-NAS (50th percentile) |
| Metric | Average HV Rank |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 17 of 17 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Average HV Rank](<https://www.sota2.com/research/sota/offline-multi-objective-optimization-on-off-moo-bench-mo-nas-50th-percentile.md?metric=Average%20HV%20Rank>): Lower is better; unit: SCALAR.

## Recorded results

| Method | Average HV Rank | Result date | Report paper |
| --- | --- | --- | --- |
| MOBO | 4.21 | Not recorded | 6a30d870d33dfb23e783cb11 |
| Multiple Models + IOM | 5.66 | Not recorded | 6a30d870d33dfb23e783cb11 |
| End-to-End | 5.94 | Not recorded | 6a30d870d33dfb23e783cb11 |
| Multi Head | 6.23 | Not recorded | 6a30d870d33dfb23e783cb11 |
| Multiple Models + RoMA | 6.24 | Not recorded | 6a30d870d33dfb23e783cb11 |
| Multiple Models | 6.73 | Not recorded | 6a30d870d33dfb23e783cb11 |
| Multiple Models + COM | 6.91 | Not recorded | 6a30d870d33dfb23e783cb11 |
| End-to-End + PcGrad | 7.56 | Not recorded | 6a30d870d33dfb23e783cb11 |
| Multi Head + PcGrad | 7.56 | Not recorded | 6a30d870d33dfb23e783cb11 |
| DOMOO | 8.36 | Not recorded | 6a30d870d33dfb23e783cb11 |
| D(best) | 9.06 | Not recorded | 6a30d870d33dfb23e783cb11 |
| Multiple Models + ICT | 9.59 | Not recorded | 6a30d870d33dfb23e783cb11 |
| Multiple Models + Tri-Mentoring | 11.1 | Not recorded | 6a30d870d33dfb23e783cb11 |
| ParetoFlow | 12.04 | Not recorded | 6a30d870d33dfb23e783cb11 |
| End-to-End + GradNorm | 12.99 | Not recorded | 6a30d870d33dfb23e783cb11 |
| MOBO-qParEGO | 13.19 | Not recorded | 6a30d870d33dfb23e783cb11 |
| Multi Head + GradNorm | 13.27 | Not recorded | 6a30d870d33dfb23e783cb11 |

### MOBO

Average HV Rank: 4.21. Result ID: 6a30fb37fbc373592c2cd3e8.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d870d33dfb23e783cb11.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa738c1e209131cea5d25f.md>)

Recorded evaluation configuration: {"Evaluation percentile":"50th"}. This record does not assert independent reproduction.

### Multiple Models + IOM

Average HV Rank: 5.66. Result ID: 6a30fb37fbc373592c2cd3e5.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d870d33dfb23e783cb11.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e091e209131cea3d97d.md>)

[Implementation](<https://sites.google.com/view/iom-neurips>)

Recorded evaluation configuration: {"Evaluation percentile":"50th"}. This record does not assert independent reproduction.

### End-to-End

Average HV Rank: 5.94. Result ID: 6a30fb37fbc373592c2cd3dc.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d870d33dfb23e783cb11.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa738c1e209131cea5d25f.md>)

Recorded evaluation configuration: {"Evaluation percentile":"50th"}. This record does not assert independent reproduction.

### Multi Head

Average HV Rank: 6.23. Result ID: 6a30fb37fbc373592c2cd3df.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d870d33dfb23e783cb11.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa738c1e209131cea5d25f.md>)

Recorded evaluation configuration: {"Evaluation percentile":"50th"}. This record does not assert independent reproduction.

### Multiple Models + RoMA

Average HV Rank: 6.24. Result ID: 6a30fb37fbc373592c2cd3e4.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d870d33dfb23e783cb11.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6b171e209131cea2c165.md>)

Recorded evaluation configuration: {"Evaluation percentile":"50th"}. This record does not assert independent reproduction.

### Multiple Models

Average HV Rank: 6.73. Result ID: 6a30fb37fbc373592c2cd3e2.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d870d33dfb23e783cb11.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa738c1e209131cea5d25f.md>)

Recorded evaluation configuration: {"Evaluation percentile":"50th"}. This record does not assert independent reproduction.

### Multiple Models + COM

Average HV Rank: 6.91. Result ID: 6a30fb37fbc373592c2cd3e3.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d870d33dfb23e783cb11.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a5f1e209131cea27cd2.md>)

Recorded evaluation configuration: {"Evaluation percentile":"50th"}. This record does not assert independent reproduction.

### End-to-End + PcGrad

Average HV Rank: 7.56. Result ID: 6a30fb37fbc373592c2cd3de.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d870d33dfb23e783cb11.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68fb1e209131cea1f828.md>)

Recorded evaluation configuration: {"Evaluation percentile":"50th"}. This record does not assert independent reproduction.

### Multi Head + PcGrad

Average HV Rank: 7.56. Result ID: 6a30fb37fbc373592c2cd3e1.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d870d33dfb23e783cb11.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68fb1e209131cea1f828.md>)

Recorded evaluation configuration: {"Evaluation percentile":"50th"}. This record does not assert independent reproduction.

### DOMOO

Average HV Rank: 8.36. Result ID: 6a30fb37fbc373592c2cd3eb.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d870d33dfb23e783cb11.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a30d870d33dfb23e783cb11.md>)

Recorded evaluation configuration: {"Evaluation percentile":"50th"}. This record does not assert independent reproduction.

### D(best)

Average HV Rank: 9.06. Result ID: 6a30fb37fbc373592c2cd3db.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d870d33dfb23e783cb11.md>)

Recorded evaluation configuration: {"Evaluation percentile":"50th"}. This record does not assert independent reproduction.

### Multiple Models + ICT

Average HV Rank: 9.59. Result ID: 6a30fb37fbc373592c2cd3e6.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d870d33dfb23e783cb11.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70cd1e209131cea4d3d1.md>)

Recorded evaluation configuration: {"Evaluation percentile":"50th"}. This record does not assert independent reproduction.

### Multiple Models + Tri-Mentoring

Average HV Rank: 11.1. Result ID: 6a30fb37fbc373592c2cd3e7.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d870d33dfb23e783cb11.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70cd1e209131cea4d3d0.md>)

Recorded evaluation configuration: {"Evaluation percentile":"50th"}. This record does not assert independent reproduction.

### ParetoFlow

Average HV Rank: 12.04. Result ID: 6a30fb37fbc373592c2cd3ea.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d870d33dfb23e783cb11.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77101e209131cea70e60.md>)

Recorded evaluation configuration: {"Evaluation percentile":"50th"}. This record does not assert independent reproduction.

### End-to-End + GradNorm

Average HV Rank: 12.99. Result ID: 6a30fb37fbc373592c2cd3dd.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d870d33dfb23e783cb11.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa649d1e209131cea0597c.md>)

Recorded evaluation configuration: {"Evaluation percentile":"50th"}. This record does not assert independent reproduction.

### MOBO-qParEGO

Average HV Rank: 13.19. Result ID: 6a30fb37fbc373592c2cd3e9.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d870d33dfb23e783cb11.md>)

Recorded evaluation configuration: {"Evaluation percentile":"50th"}. This record does not assert independent reproduction.

### Multi Head + GradNorm

Average HV Rank: 13.27. Result ID: 6a30fb37fbc373592c2cd3e0.

[Reporting paper](<https://www.sota2.com/research/paper/6a30d870d33dfb23e783cb11.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa649d1e209131cea0597c.md>)

Recorded evaluation configuration: {"Evaluation percentile":"50th"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
