# Image Classification on CIFAR-100 Long-Tailed (rho=50) (test)

Source: [Image Classification on CIFAR-100 Long-Tailed (rho=50) (test)](<https://www.sota2.com/research/sota/image-classification-on-cifar-100-long-tailed-rho-50-test>)

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

Data updated: 2026-05-19T11:41:20.900Z

Evaluation of Image Classification performance on the test split of the CIFAR-100 Long-Tailed dataset configured with an imbalance ratio (rho) of 50, reporting Top-1 error rate.

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 | Image Classification |
| Dataset (including any stated split) | CIFAR-100 Long-Tailed (rho=50) (test) |
| Metric | Top-1 Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 44 of 44 |
| 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

- [Top-1 Accuracy](<https://www.sota2.com/research/sota/image-classification-on-cifar-100-long-tailed-rho-50-test.md?metric=Top-1%20Accuracy>): Higher is better; unit: PERCENT.

- [Top-1 Error Rate](<https://www.sota2.com/research/sota/image-classification-on-cifar-100-long-tailed-rho-50-test.md?metric=Top-1%20Error%20Rate>): Lower is better; unit: PERCENT.

## Recorded results

| Method | Top-1 Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| DeiT-LT | 60.5 | Not recorded | 68fa72d21e209131cea58b99 |
| PaCo+SAM | 56.6 | Not recorded | 68fa72d21e209131cea58b99 |
| DeiT-LT | 54.1 | Not recorded | 68fa72d21e209131cea58b99 |
| GCL | 53.6 | Not recorded | 68fa72d21e209131cea58b99 |
| MISLAS | 52.3 | Not recorded | 68fa72d21e209131cea58b99 |
| Hybrid-SC | 51.9 | Not recorded | 68fa72d21e209131cea58b99 |
| ACE | 51.9 | Not recorded | 68fa72d21e209131cea58b99 |
| BiKD | 51.84 | Not recorded | 6a0beef1dbc836b87d82c901 |
| CAM | 51.7 | Not recorded | 68fa72d21e209131cea58b99 |
| Teacher(CE) | 51.59 | Not recorded | 6a0beef1dbc836b87d82c901 |
| DiVE | 51.01 | Not recorded | 6a0beef1dbc836b87d82c901 |
| BKD | 50.71 | Not recorded | 6a0beef1dbc836b87d82c901 |
| SSD | 50.5 | Not recorded | 68fa72d21e209131cea58b99 |
| L-D-SAM | 49.4 | Not recorded | 68fa72d21e209131cea58b99 |
| LFME | 49.29 | Not recorded | 6a0beef1dbc836b87d82c901 |
| LDAM+DAP | 49.2 | Not recorded | 68fa72d21e209131cea58b99 |
| CCKD | 48.86 | Not recorded | 6a0beef1dbc836b87d82c901 |
| Student(CE) | 48.14 | Not recorded | 6a0beef1dbc836b87d82c901 |
| KD | 47.87 | Not recorded | 6a0beef1dbc836b87d82c901 |
| RWKD | 47.26 | Not recorded | 6a0beef1dbc836b87d82c901 |
| WSL | 47.2 | Not recorded | 6a0beef1dbc836b87d82c901 |
| BBN | 47 | Not recorded | 68fa72d21e209131cea58b99 |
| CB Focal loss | 46.2 | Not recorded | 68fa72d21e209131cea58b99 |
| LDAM+DRW | 45.1 | Not recorded | 68fa72d21e209131cea58b99 |
| DeiT-III | 44.1 | Not recorded | 68fa72d21e209131cea58b99 |
| ViT (cRT) | 42.2 | Not recorded | 68fa72d21e209131cea58b99 |
| DeiT | 39.1 | Not recorded | 68fa72d21e209131cea58b99 |
| ViT | 39 | Not recorded | 68fa72d21e209131cea58b99 |

### DeiT-LT

Top-1 Accuracy: 60.5. Result ID: 69929a5f3a6c314849101461.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72d21e209131cea58b99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72d21e209131cea58b99.md>)

[Implementation](<https://rangwani-harsh.github.io/DeiT-LT>)

Recorded evaluation configuration: {"Backbone":"ViT-B","Teacher":"PaCo+SAM"}. This record does not assert independent reproduction.

### PaCo+SAM

Top-1 Accuracy: 56.6. Result ID: 69929a5f3a6c31484910145b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72d21e209131cea58b99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e491e209131cea3f009.md>)

[Implementation](<https://github.com/val-iisc/Saddle-LongTail>)

Recorded evaluation configuration: {"Backbone":"ResNet32"}. This record does not assert independent reproduction.

### DeiT-LT

Top-1 Accuracy: 54.1. Result ID: 69929a5f3a6c314849101460.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72d21e209131cea58b99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72d21e209131cea58b99.md>)

[Implementation](<https://rangwani-harsh.github.io/DeiT-LT>)

Recorded evaluation configuration: {"Backbone":"ViT-B","Teacher":"L-D-SAM"}. This record does not assert independent reproduction.

### GCL

Top-1 Accuracy: 53.6. Result ID: 69929a5f3a6c314849101459.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72d21e209131cea58b99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f711e209131cea459b7.md>)

[Implementation](<https://github.com/Keke921/GCLLoss>)

Recorded evaluation configuration: {"Backbone":"ResNet32"}. This record does not assert independent reproduction.

### MISLAS

Top-1 Accuracy: 52.3. Result ID: 69929a5f3a6c314849101455.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72d21e209131cea58b99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69b21e209131cea237df.md>)

[Implementation](<https://github.com/Jia-Research-Lab/MiSLAS>)

Recorded evaluation configuration: {"Backbone":"ResNet32"}. This record does not assert independent reproduction.

### Hybrid-SC

Top-1 Accuracy: 51.9. Result ID: 69929a5f3a6c314849101456.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72d21e209131cea58b99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa699d1e209131cea23186.md>)

[Implementation](<https://k-han.github.io/HybridLT>)

Recorded evaluation configuration: {"Backbone":"ResNet32"}. This record does not assert independent reproduction.

### ACE

Top-1 Accuracy: 51.9. Result ID: 69929a5f3a6c314849101458.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72d21e209131cea58b99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a7e1e209131cea2899b.md>)

[Implementation](<https://github.com/jrcai/ACE>)

Recorded evaluation configuration: {"Backbone":"ResNet32"}. This record does not assert independent reproduction.

### BiKD

Top-1 Accuracy: 51.84. Result ID: 6a0c4c5cfbc373592c03709b.

[Reporting paper](<https://www.sota2.com/research/paper/6a0beef1dbc836b87d82c901.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a0beef1dbc836b87d82c901.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### CAM

Top-1 Accuracy: 51.7. Result ID: 69929a5f3a6c314849101454.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72d21e209131cea58b99.md>)

Recorded evaluation configuration: {"Backbone":"ResNet32"}. This record does not assert independent reproduction.

### Teacher(CE)

Top-1 Accuracy: 51.59. Result ID: 6a0c4c5cfbc373592c037092.

[Reporting paper](<https://www.sota2.com/research/paper/6a0beef1dbc836b87d82c901.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### DiVE

Top-1 Accuracy: 51.01. Result ID: 6a0c4c5cfbc373592c03709a.

[Reporting paper](<https://www.sota2.com/research/paper/6a0beef1dbc836b87d82c901.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6acb1e209131cea2a4a4.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### BKD

Top-1 Accuracy: 50.71. Result ID: 6a0c4c5cfbc373592c037095.

[Reporting paper](<https://www.sota2.com/research/paper/6a0beef1dbc836b87d82c901.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69ca1e209131cea2403a.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### SSD

Top-1 Accuracy: 50.5. Result ID: 69929a5f3a6c314849101457.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72d21e209131cea58b99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ab91e209131cea2a0fa.md>)

[Implementation](<https://github.com/MCG-NJU/SSD-LT>)

Recorded evaluation configuration: {"Backbone":"ResNet32"}. This record does not assert independent reproduction.

### L-D-SAM

Top-1 Accuracy: 49.4. Result ID: 69929a5f3a6c31484910145a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72d21e209131cea58b99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e491e209131cea3f009.md>)

[Implementation](<https://github.com/val-iisc/Saddle-LongTail>)

Recorded evaluation configuration: {"Backbone":"ResNet32"}. This record does not assert independent reproduction.

### LFME

Top-1 Accuracy: 49.29. Result ID: 6a0c4c5cfbc373592c037099.

[Reporting paper](<https://www.sota2.com/research/paper/6a0beef1dbc836b87d82c901.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68681e209131cea1beb9.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### LDAM+DAP

Top-1 Accuracy: 49.2. Result ID: 69929a5f3a6c314849101452.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72d21e209131cea58b99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67401e209131cea1539a.md>)

Recorded evaluation configuration: {"Backbone":"ResNet32"}. This record does not assert independent reproduction.

### CCKD

Top-1 Accuracy: 48.86. Result ID: 6a0c4c5cfbc373592c037096.

[Reporting paper](<https://www.sota2.com/research/paper/6a0beef1dbc836b87d82c901.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Student(CE)

Top-1 Accuracy: 48.14. Result ID: 6a0c4c5cfbc373592c037093.

[Reporting paper](<https://www.sota2.com/research/paper/6a0beef1dbc836b87d82c901.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### KD

Top-1 Accuracy: 47.87. Result ID: 6a0c4c5cfbc373592c037094.

[Reporting paper](<https://www.sota2.com/research/paper/6a0beef1dbc836b87d82c901.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa61b81e209131ce9f5b87.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### RWKD

Top-1 Accuracy: 47.26. Result ID: 6a0c4c5cfbc373592c037098.

[Reporting paper](<https://www.sota2.com/research/paper/6a0beef1dbc836b87d82c901.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### WSL

Top-1 Accuracy: 47.2. Result ID: 6a0c4c5cfbc373592c037097.

[Reporting paper](<https://www.sota2.com/research/paper/6a0beef1dbc836b87d82c901.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa692c1e209131cea208d6.md>)

[Implementation](<https://github.com/bellymonster/Weighted-Soft-Label-Distillation>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### BBN

Top-1 Accuracy: 47. Result ID: 69929a5f3a6c314849101453.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72d21e209131cea58b99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67371e209131cea14e59.md>)

[Implementation](<https://github.com/Megvii-Nanjing/BBN>)

Recorded evaluation configuration: {"Backbone":"ResNet32"}. This record does not assert independent reproduction.

### CB Focal loss

Top-1 Accuracy: 46.2. Result ID: 69929a5f3a6c314849101450.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72d21e209131cea58b99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa654f1e209131cea0988a.md>)

[Implementation](<https://github.com/richardaecn/class-balanced-loss>)

Recorded evaluation configuration: {"Backbone":"ResNet32"}. This record does not assert independent reproduction.

### LDAM+DRW

Top-1 Accuracy: 45.1. Result ID: 69929a5f3a6c314849101451.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72d21e209131cea58b99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66901e209131cea10e0f.md>)

[Implementation](<https://github.com/kaidic/LDAM-DRW>)

Recorded evaluation configuration: {"Backbone":"ResNet32"}. This record does not assert independent reproduction.

### DeiT-III

Top-1 Accuracy: 44.1. Result ID: 69929a5f3a6c31484910145f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72d21e209131cea58b99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6c591e209131cea3378d.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B"}. This record does not assert independent reproduction.

### ViT (cRT)

Top-1 Accuracy: 42.2. Result ID: 69929a5f3a6c31484910145d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72d21e209131cea58b99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa671b1e209131cea144a9.md>)

[Implementation](<https://github.com/facebookresearch/classifier-balancing>)

Recorded evaluation configuration: {"Backbone":"ViT-B"}. This record does not assert independent reproduction.

### DeiT

Top-1 Accuracy: 39.1. Result ID: 69929a5f3a6c31484910145e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72d21e209131cea58b99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69181e209131cea202b4.md>)

[Implementation](<https://github.com/facebookresearch/deit>)

Recorded evaluation configuration: {"Backbone":"ViT-B"}. This record does not assert independent reproduction.

### ViT

Top-1 Accuracy: 39. Result ID: 69929a5f3a6c31484910145c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa72d21e209131cea58b99.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6a0b1e209131cea25b04.md>)

[Implementation](<https://github.com/google-research/vision_transformer>)

Recorded evaluation configuration: {"Backbone":"ViT-B"}. This record does not assert independent reproduction.

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