# Predicate Classification on PSG 1.0 (test)

Source: [Predicate Classification on PSG 1.0 (test)](<https://www.sota2.com/research/sota/predicate-classification-on-psg-1-0-test>)

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

Data updated: 2026-02-26T05:59:40.914Z

Evaluation of Predicate Classification performance on the test split of the PSG 1.0 dataset, measured using R@K and mR@K metrics.

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 | Predicate Classification |
| Dataset (including any stated split) | PSG 1.0 (test) |
| Metric | R@20 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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

- [R@20](<https://www.sota2.com/research/sota/predicate-classification-on-psg-1-0-test.md?metric=R%4020>): Higher is better; unit: PERCENT.

- [mR@20](<https://www.sota2.com/research/sota/predicate-classification-on-psg-1-0-test.md?metric=mR%4020>): Higher is better; unit: PERCENT.

- [R@50](<https://www.sota2.com/research/sota/predicate-classification-on-psg-1-0-test.md?metric=R%4050>): Higher is better; unit: PERCENT.

- [mR@50](<https://www.sota2.com/research/sota/predicate-classification-on-psg-1-0-test.md?metric=mR%4050>): Higher is better; unit: PERCENT.

- [R@100](<https://www.sota2.com/research/sota/predicate-classification-on-psg-1-0-test.md?metric=R%40100>): Higher is better; unit: PERCENT.

- [mR@100](<https://www.sota2.com/research/sota/predicate-classification-on-psg-1-0-test.md?metric=mR%40100>): Higher is better; unit: PERCENT.

## Recorded results

| Method | R@20 | Result date | Report paper |
| --- | --- | --- | --- |
| VCTree | 45.9 | Not recorded | 68fa6d131e209131cea37790 |
| VCTree | 45.3 | Not recorded | 68fa6d131e209131cea37790 |
| MOTIFS | 45.1 | Not recorded | 68fa6d131e209131cea37790 |
| MOTIFS | 44.9 | Not recorded | 68fa6d131e209131cea37790 |
| GPSNet | 38.8 | Not recorded | 68fa6d131e209131cea37790 |
| IMP | 31.9 | Not recorded | 68fa6d131e209131cea37790 |
| GPSNet | 31.5 | Not recorded | 68fa6d131e209131cea37790 |
| IMP | 30.5 | Not recorded | 68fa6d131e209131cea37790 |

### VCTree

R@20: 45.9. Result ID: 698f7a5e3a6c314849f90e59.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d131e209131cea37790.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa651d1e209131cea086da.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-101","Training Epochs":12}. This record does not assert independent reproduction.

### VCTree

R@20: 45.3. Result ID: 698f7a5e3a6c314849f90e55.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d131e209131cea37790.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa651d1e209131cea086da.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-50","Training Epochs":12}. This record does not assert independent reproduction.

### MOTIFS

R@20: 45.1. Result ID: 698f7a5e3a6c314849f90e58.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d131e209131cea37790.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa642f1e209131cea034c6.md>)

[Implementation](<https://github.com/rowanz/neural-motifs>)

Recorded evaluation configuration: {"Backbone":"ResNet-101","Training Epochs":12}. This record does not assert independent reproduction.

### MOTIFS

R@20: 44.9. Result ID: 698f7a5e3a6c314849f90e54.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d131e209131cea37790.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa642f1e209131cea034c6.md>)

[Implementation](<https://github.com/rowanz/neural-motifs>)

Recorded evaluation configuration: {"Backbone":"ResNet-50","Training Epochs":12}. This record does not assert independent reproduction.

### GPSNet

R@20: 38.8. Result ID: 698f7a5e3a6c314849f90e5a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d131e209131cea37790.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa673d1e209131cea151d0.md>)

[Implementation](<https://github.com/taksau/GPS-Net>)

Recorded evaluation configuration: {"Backbone":"ResNet-101","Training Epochs":12}. This record does not assert independent reproduction.

### IMP

R@20: 31.9. Result ID: 698f7a5e3a6c314849f90e53.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d131e209131cea37790.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa634d1e209131ce9fde50.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-50","Training Epochs":12}. This record does not assert independent reproduction.

### GPSNet

R@20: 31.5. Result ID: 698f7a5e3a6c314849f90e56.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d131e209131cea37790.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa673d1e209131cea151d0.md>)

[Implementation](<https://github.com/taksau/GPS-Net>)

Recorded evaluation configuration: {"Backbone":"ResNet-50","Training Epochs":12}. This record does not assert independent reproduction.

### IMP

R@20: 30.5. Result ID: 698f7a5e3a6c314849f90e57.

[Reporting paper](<https://www.sota2.com/research/paper/68fa6d131e209131cea37790.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa634d1e209131ce9fde50.md>)

Recorded evaluation configuration: {"Backbone":"ResNet-101","Training Epochs":12}. This record does not assert independent reproduction.

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