# Fine-grained Image Classification on AwA2 Bobcat vs. Leopard (test)

Source: [Fine-grained Image Classification on AwA2 Bobcat vs. Leopard (test)](<https://www.sota2.com/research/sota/fine-grained-image-classification-on-awa2-bobcat-vs-leopard-test>)

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

Data updated: 2026-07-07T11:02:25.649Z

Evaluation of Fine-grained Image Classification performance on a specific subset of classes (Bobcat vs. Leopard) within the test split of the Animals with Attributes 2 (AwA2) dataset, utilizing specialized metrics like concept cosine similarity and LP/BC accuracy gaps.

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 | Fine-grained Image Classification |
| Dataset (including any stated split) | AwA2 Bobcat vs. Leopard (test) |
| Metric | Avg Acc (BC) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| 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

- [Avg Acc (BC)](<https://www.sota2.com/research/sota/fine-grained-image-classification-on-awa2-bobcat-vs-leopard-test.md?metric=Avg%20Acc%20(BC)>): Higher is better; unit: PERCENT.

- [Avg Cross-Model Concept Cos Sim](<https://www.sota2.com/research/sota/fine-grained-image-classification-on-awa2-bobcat-vs-leopard-test.md?metric=Avg%20Cross-Model%20Concept%20Cos%20Sim>): Higher is better; unit: SCALAR.

- [Avg Acc (LP)](<https://www.sota2.com/research/sota/fine-grained-image-classification-on-awa2-bobcat-vs-leopard-test.md?metric=Avg%20Acc%20(LP)>): Higher is better; unit: PERCENT.

- [Avg Acc Gap (LP-BC)](<https://www.sota2.com/research/sota/fine-grained-image-classification-on-awa2-bobcat-vs-leopard-test.md?metric=Avg%20Acc%20Gap%20(LP-BC)>): Lower is better; unit: PERCENT.

## Recorded results

| Method | Avg Acc (BC) | Result date | Report paper |
| --- | --- | --- | --- |
| Rashomon-Slice | 98.17 | Not recorded | 692689e8d6b0dd33429306a2 |
| Random Init | 94.21 | Not recorded | 692689e8d6b0dd33429306a2 |
| Dropout | 93.65 | Not recorded | 692689e8d6b0dd33429306a2 |
| DivEnsemble | 93.1 | Not recorded | 692689e8d6b0dd33429306a2 |

### Rashomon-Slice

Avg Acc (BC): 98.17. Result ID: 6a4cdcaf60b88bb339cb2589.

[Reporting paper](<https://www.sota2.com/research/paper/692689e8d6b0dd33429306a2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/692689e8d6b0dd33429306a2.md>)

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

### Random Init

Avg Acc (BC): 94.21. Result ID: 6a4cdcaf60b88bb339cb2586.

[Reporting paper](<https://www.sota2.com/research/paper/692689e8d6b0dd33429306a2.md>)

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

### Dropout

Avg Acc (BC): 93.65. Result ID: 6a4cdcaf60b88bb339cb2587.

[Reporting paper](<https://www.sota2.com/research/paper/692689e8d6b0dd33429306a2.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa720f1e209131cea543eb.md>)

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

### DivEnsemble

Avg Acc (BC): 93.1. Result ID: 6a4cdcaf60b88bb339cb2588.

[Reporting paper](<https://www.sota2.com/research/paper/692689e8d6b0dd33429306a2.md>)

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

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