# Ranking past classes by forgetting on CIFAR-100 (incremental)

Source: [Ranking past classes by forgetting on CIFAR-100 (incremental)](<https://www.sota2.com/research/sota/ranking-past-classes-by-forgetting-on-cifar-100-incremental>)

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

Data updated: 2026-05-15T09:20:05.993Z

Evaluation of model forgetting behavior on the CIFAR-100 dataset under an incremental learning setting, specifically measuring the correlation (e.g., Spearman's rho) between predicted forgetting and actual forgetting across past classes.

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 | Ranking past classes by forgetting |
| Dataset (including any stated split) | CIFAR-100 (incremental) |
| Metric | Spearman's Rho |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 9 of 9 |
| 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

- [Spearman's Rho](<https://www.sota2.com/research/sota/ranking-past-classes-by-forgetting-on-cifar-100-incremental.md?metric=Spearman%27s%20Rho>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Spearman's Rho | Result date | Report paper |
| --- | --- | --- | --- |
| Last-Layer Imbalanced Forgetting Coefficients | 0.91 | Not recorded | 6a06a8b8dbc836b87d82a204 |
| Last-Layer Imbalanced Forgetting Coefficients | 0.9 | Not recorded | 6a06a8b8dbc836b87d82a204 |
| Last-Layer Imbalanced Forgetting Coefficients | 0.88 | Not recorded | 6a06a8b8dbc836b87d82a204 |
| New-Dataset Interference Coefficient (NIC) | 0.75 | Not recorded | 6a06a8b8dbc836b87d82a204 |
| New-Dataset Interference Coefficient (NIC) | 0.74 | Not recorded | 6a06a8b8dbc836b87d82a204 |
| LOG-SIM | 0.67 | Not recorded | 6a06a8b8dbc836b87d82a204 |
| New-Dataset Interference Coefficient (NIC) | 0.65 | Not recorded | 6a06a8b8dbc836b87d82a204 |
| LOG-SIM | 0.63 | Not recorded | 6a06a8b8dbc836b87d82a204 |
| LOG-SIM | 0.39 | Not recorded | 6a06a8b8dbc836b87d82a204 |

### Last-Layer Imbalanced Forgetting Coefficients

Spearman's Rho: 0.91. Result ID: 6a06e53547f34ce924eed69d.

[Reporting paper](<https://www.sota2.com/research/paper/6a06a8b8dbc836b87d82a204.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a06a8b8dbc836b87d82a204.md>)

Recorded evaluation configuration: {"Incremental Step":"2","Subset":"{10%, 20%}","Evaluation Protocol":"SW-LOO"}. This record does not assert independent reproduction.

### Last-Layer Imbalanced Forgetting Coefficients

Spearman's Rho: 0.9. Result ID: 6a06e53547f34ce924eed69c.

[Reporting paper](<https://www.sota2.com/research/paper/6a06a8b8dbc836b87d82a204.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a06a8b8dbc836b87d82a204.md>)

Recorded evaluation configuration: {"Incremental Step":"2","Evaluation Protocol":"SW-LOO"}. This record does not assert independent reproduction.

### Last-Layer Imbalanced Forgetting Coefficients

Spearman's Rho: 0.88. Result ID: 6a06e53547f34ce924eed69e.

[Reporting paper](<https://www.sota2.com/research/paper/6a06a8b8dbc836b87d82a204.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a06a8b8dbc836b87d82a204.md>)

Recorded evaluation configuration: {"Incremental Step":"3","Evaluation Protocol":"SW-LOO"}. This record does not assert independent reproduction.

### New-Dataset Interference Coefficient (NIC)

Spearman's Rho: 0.75. Result ID: 6a06e53547f34ce924eed6a0.

[Reporting paper](<https://www.sota2.com/research/paper/6a06a8b8dbc836b87d82a204.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a06a8b8dbc836b87d82a204.md>)

Recorded evaluation configuration: {"Incremental Step":"2","Subset":"{10%, 20%}"}. This record does not assert independent reproduction.

### New-Dataset Interference Coefficient (NIC)

Spearman's Rho: 0.74. Result ID: 6a06e53547f34ce924eed69f.

[Reporting paper](<https://www.sota2.com/research/paper/6a06a8b8dbc836b87d82a204.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a06a8b8dbc836b87d82a204.md>)

Recorded evaluation configuration: {"Incremental Step":"2"}. This record does not assert independent reproduction.

### LOG-SIM

Spearman's Rho: 0.67. Result ID: 6a06e53547f34ce924eed6a3.

[Reporting paper](<https://www.sota2.com/research/paper/6a06a8b8dbc836b87d82a204.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72a01e209131cea57a46.md>)

Recorded evaluation configuration: {"Incremental Step":"2","Subset":"{10%, 20%}"}. This record does not assert independent reproduction.

### New-Dataset Interference Coefficient (NIC)

Spearman's Rho: 0.65. Result ID: 6a06e53547f34ce924eed6a1.

[Reporting paper](<https://www.sota2.com/research/paper/6a06a8b8dbc836b87d82a204.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a06a8b8dbc836b87d82a204.md>)

Recorded evaluation configuration: {"Incremental Step":"3"}. This record does not assert independent reproduction.

### LOG-SIM

Spearman's Rho: 0.63. Result ID: 6a06e53547f34ce924eed6a2.

[Reporting paper](<https://www.sota2.com/research/paper/6a06a8b8dbc836b87d82a204.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72a01e209131cea57a46.md>)

Recorded evaluation configuration: {"Incremental Step":"2"}. This record does not assert independent reproduction.

### LOG-SIM

Spearman's Rho: 0.39. Result ID: 6a06e53547f34ce924eed6a4.

[Reporting paper](<https://www.sota2.com/research/paper/6a06a8b8dbc836b87d82a204.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72a01e209131cea57a46.md>)

Recorded evaluation configuration: {"Incremental Step":"3"}. This record does not assert independent reproduction.

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