# Generalized Category Discovery on MSTAR Inductive

Source: [Generalized Category Discovery on MSTAR Inductive](<https://www.sota2.com/research/sota/generalized-category-discovery-on-mstar-inductive>)

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

Data updated: 2026-04-27T06:59:35.785Z

Evaluation of Generalized Category Discovery performance on the MSTAR Inductive dataset, reporting metrics for all categories, known categories ('Old'), and novel categories ('New').

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 | Generalized Category Discovery |
| Dataset (including any stated split) | MSTAR Inductive |
| Metric | Accuracy (All) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 15 of 15 |
| 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

- [Accuracy (All)](<https://www.sota2.com/research/sota/generalized-category-discovery-on-mstar-inductive.md?metric=Accuracy%20(All)>): Higher is better; unit: PERCENT.

- [Accuracy (Old)](<https://www.sota2.com/research/sota/generalized-category-discovery-on-mstar-inductive.md?metric=Accuracy%20(Old)>): Higher is better; unit: PERCENT.

- [Accuracy (New)](<https://www.sota2.com/research/sota/generalized-category-discovery-on-mstar-inductive.md?metric=Accuracy%20(New)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy (All) | Result date | Report paper |
| --- | --- | --- | --- |
| MCPT | 97.98 | Not recorded | 69eeed9255a3c32af89f977d |
| MCPT | 93.36 | Not recorded | 69eeed9255a3c32af89f977d |
| ProtoGCD (w/ CM-PT) | 91.38 | Not recorded | 69eeed9255a3c32af89f977d |
| ProtoGCD | 90.14 | Not recorded | 69eeed9255a3c32af89f977d |
| SimGCD | 89.44 | Not recorded | 69eeed9255a3c32af89f977d |
| BKD-CL | 87.3 | Not recorded | 69eeed9255a3c32af89f977d |
| InfoSieve | 85.9 | Not recorded | 69eeed9255a3c32af89f977d |
| CMS | 85.57 | Not recorded | 69eeed9255a3c32af89f977d |
| ProtoGCD | 84.95 | Not recorded | 69eeed9255a3c32af89f977d |
| GCD | 84.54 | Not recorded | 69eeed9255a3c32af89f977d |
| BKD-CL | 84.04 | Not recorded | 69eeed9255a3c32af89f977d |
| CMS | 81.15 | Not recorded | 69eeed9255a3c32af89f977d |
| SimGCD | 79.71 | Not recorded | 69eeed9255a3c32af89f977d |
| GCD | 75.46 | Not recorded | 69eeed9255a3c32af89f977d |
| InfoSieve | 75.05 | Not recorded | 69eeed9255a3c32af89f977d |

### MCPT

Accuracy (All): 97.98. Result ID: 69ef07a347f34ce924d4a472.

[Reporting paper](<https://www.sota2.com/research/paper/69eeed9255a3c32af89f977d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69eeed9255a3c32af89f977d.md>)

Recorded evaluation configuration: {"Backbone":"DINOv2","Evaluation Protocol":"Inductive"}. This record does not assert independent reproduction.

### MCPT

Accuracy (All): 93.36. Result ID: 69ef07a347f34ce924d4a46a.

[Reporting paper](<https://www.sota2.com/research/paper/69eeed9255a3c32af89f977d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69eeed9255a3c32af89f977d.md>)

Recorded evaluation configuration: {"Backbone":"DINOv1","Evaluation Protocol":"Inductive"}. This record does not assert independent reproduction.

### ProtoGCD (w/ CM-PT)

Accuracy (All): 91.38. Result ID: 69ef07a347f34ce924d4a471.

[Reporting paper](<https://www.sota2.com/research/paper/69eeed9255a3c32af89f977d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69eeed9255a3c32af89f977d.md>)

Recorded evaluation configuration: {"Backbone":"DINOv2","Evaluation Protocol":"Inductive"}. This record does not assert independent reproduction.

### ProtoGCD

Accuracy (All): 90.14. Result ID: 69ef07a347f34ce924d4a470.

[Reporting paper](<https://www.sota2.com/research/paper/69eeed9255a3c32af89f977d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77e21e209131cea75cd4.md>)

Recorded evaluation configuration: {"Backbone":"DINOv2","Evaluation Protocol":"Inductive"}. This record does not assert independent reproduction.

### SimGCD

Accuracy (All): 89.44. Result ID: 69ef07a347f34ce924d4a46c.

[Reporting paper](<https://www.sota2.com/research/paper/69eeed9255a3c32af89f977d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71741e209131cea51542.md>)

[Implementation](<https://github.com/CVMI-Lab/SimGCD>)

Recorded evaluation configuration: {"Backbone":"DINOv2","Evaluation Protocol":"Inductive"}. This record does not assert independent reproduction.

### BKD-CL

Accuracy (All): 87.3. Result ID: 69ef07a347f34ce924d4a46f.

[Reporting paper](<https://www.sota2.com/research/paper/69eeed9255a3c32af89f977d.md>)

Recorded evaluation configuration: {"Backbone":"DINOv2","Evaluation Protocol":"Inductive"}. This record does not assert independent reproduction.

### InfoSieve

Accuracy (All): 85.9. Result ID: 69ef07a347f34ce924d4a46d.

[Reporting paper](<https://www.sota2.com/research/paper/69eeed9255a3c32af89f977d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa5b4d1e209131ce9c70c7.md>)

[Implementation](<https://github.com/SarahRastegar/InfoSieve>)

Recorded evaluation configuration: {"Backbone":"DINOv2","Evaluation Protocol":"Inductive"}. This record does not assert independent reproduction.

### CMS

Accuracy (All): 85.57. Result ID: 69ef07a347f34ce924d4a46e.

[Reporting paper](<https://www.sota2.com/research/paper/69eeed9255a3c32af89f977d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72ed1e209131cea5949c.md>)

Recorded evaluation configuration: {"Backbone":"DINOv2","Evaluation Protocol":"Inductive"}. This record does not assert independent reproduction.

### ProtoGCD

Accuracy (All): 84.95. Result ID: 69ef07a347f34ce924d4a469.

[Reporting paper](<https://www.sota2.com/research/paper/69eeed9255a3c32af89f977d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77e21e209131cea75cd4.md>)

Recorded evaluation configuration: {"Backbone":"DINOv1","Evaluation Protocol":"Inductive"}. This record does not assert independent reproduction.

### GCD

Accuracy (All): 84.54. Result ID: 69ef07a347f34ce924d4a46b.

[Reporting paper](<https://www.sota2.com/research/paper/69eeed9255a3c32af89f977d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cd91e209131cea36416.md>)

[Implementation](<https://www.robots.ox.ac.uk/~vgg/research/gcd>)

Recorded evaluation configuration: {"Backbone":"DINOv2","Evaluation Protocol":"Inductive"}. This record does not assert independent reproduction.

### BKD-CL

Accuracy (All): 84.04. Result ID: 69ef07a347f34ce924d4a468.

[Reporting paper](<https://www.sota2.com/research/paper/69eeed9255a3c32af89f977d.md>)

Recorded evaluation configuration: {"Backbone":"DINOv1","Evaluation Protocol":"Inductive"}. This record does not assert independent reproduction.

### CMS

Accuracy (All): 81.15. Result ID: 69ef07a347f34ce924d4a467.

[Reporting paper](<https://www.sota2.com/research/paper/69eeed9255a3c32af89f977d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72ed1e209131cea5949c.md>)

Recorded evaluation configuration: {"Backbone":"DINOv1","Evaluation Protocol":"Inductive"}. This record does not assert independent reproduction.

### SimGCD

Accuracy (All): 79.71. Result ID: 69ef07a347f34ce924d4a465.

[Reporting paper](<https://www.sota2.com/research/paper/69eeed9255a3c32af89f977d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71741e209131cea51542.md>)

[Implementation](<https://github.com/CVMI-Lab/SimGCD>)

Recorded evaluation configuration: {"Backbone":"DINOv1","Evaluation Protocol":"Inductive"}. This record does not assert independent reproduction.

### GCD

Accuracy (All): 75.46. Result ID: 69ef07a347f34ce924d4a464.

[Reporting paper](<https://www.sota2.com/research/paper/69eeed9255a3c32af89f977d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cd91e209131cea36416.md>)

[Implementation](<https://www.robots.ox.ac.uk/~vgg/research/gcd>)

Recorded evaluation configuration: {"Backbone":"DINOv1","Evaluation Protocol":"Inductive"}. This record does not assert independent reproduction.

### InfoSieve

Accuracy (All): 75.05. Result ID: 69ef07a347f34ce924d4a466.

[Reporting paper](<https://www.sota2.com/research/paper/69eeed9255a3c32af89f977d.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa5b4d1e209131ce9c70c7.md>)

[Implementation](<https://github.com/SarahRastegar/InfoSieve>)

Recorded evaluation configuration: {"Backbone":"DINOv1","Evaluation Protocol":"Inductive"}. This record does not assert independent reproduction.

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