# Image Classification on ImageNet 1K (val) (Accuracy vs Baselines)

Source: [Image Classification on ImageNet 1K (val) (Accuracy vs Baselines)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-1k-val-accuracy-vs-baselines>)

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

Data updated: 2026-03-16T08:41:11.267Z

Evaluation of Image Classification performance on the validation split of ImageNet 1K, reporting standard accuracy alongside comparisons against Random and Wanda baselines.

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) | ImageNet 1K (val) |
| Metric | Accuracy (%) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 6 of 6 |
| 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 (%)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-1k-val-accuracy-vs-baselines.md?metric=Accuracy%20(%25)>): Higher is better; unit: PERCENT.

- [Accuracy vs. Random](<https://www.sota2.com/research/sota/image-classification-on-imagenet-1k-val-accuracy-vs-baselines.md?metric=Accuracy%20vs.%20Random>): Higher is better; unit: SCALAR.

- [Accuracy vs. Wanda](<https://www.sota2.com/research/sota/image-classification-on-imagenet-1k-val-accuracy-vs-baselines.md?metric=Accuracy%20vs.%20Wanda>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Accuracy (%) | Result date | Report paper |
| --- | --- | --- | --- |
| Unpruned Baseline | 80.35 | Not recorded | 69b78e930636e5bf4afbdc23 |
| AGF Pruning | 64.99 | Not recorded | 69b78e930636e5bf4afbdc23 |
| Random Pruning | 64.93 | Not recorded | 69b78e930636e5bf4afbdc23 |
| Wanda Pruning | 64.69 | Not recorded | 69b78e930636e5bf4afbdc23 |
| ℓ1-Norm Pruning | 64.54 | Not recorded | 69b78e930636e5bf4afbdc23 |
| Taylor Pruning | 64.42 | Not recorded | 69b78e930636e5bf4afbdc23 |

### Unpruned Baseline

Accuracy (%): 80.35. Result ID: 69b7c21047f34ce9249f7006.

[Reporting paper](<https://www.sota2.com/research/paper/69b78e930636e5bf4afbdc23.md>)

Recorded evaluation configuration: {"Metric Proxy":"Full Capacity","Capacity":"100%","Backbone":"ResNet-50"}. This record does not assert independent reproduction.

### AGF Pruning

Accuracy (%): 64.99. Result ID: 69b7c21047f34ce9249f700b.

[Reporting paper](<https://www.sota2.com/research/paper/69b78e930636e5bf4afbdc23.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b78e930636e5bf4afbdc23.md>)

Recorded evaluation configuration: {"Metric Proxy":"Feature Sensitivity","Capacity":"25%","Backbone":"ResNet-50"}. This record does not assert independent reproduction.

### Random Pruning

Accuracy (%): 64.93. Result ID: 69b7c21047f34ce9249f700a.

[Reporting paper](<https://www.sota2.com/research/paper/69b78e930636e5bf4afbdc23.md>)

Recorded evaluation configuration: {"Metric Proxy":"Uniform Sampling","Capacity":"25%","Backbone":"ResNet-50"}. This record does not assert independent reproduction.

### Wanda Pruning

Accuracy (%): 64.69. Result ID: 69b7c21047f34ce9249f7009.

[Reporting paper](<https://www.sota2.com/research/paper/69b78e930636e5bf4afbdc23.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73281e209131cea5ac45.md>)

Recorded evaluation configuration: {"Metric Proxy":"Weights × Act","Capacity":"25%","Backbone":"ResNet-50"}. This record does not assert independent reproduction.

### ℓ1-Norm Pruning

Accuracy (%): 64.54. Result ID: 69b7c21047f34ce9249f7008.

[Reporting paper](<https://www.sota2.com/research/paper/69b78e930636e5bf4afbdc23.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa632f1e209131ce9fd215.md>)

Recorded evaluation configuration: {"Metric Proxy":"Weights (\|W\|)","Capacity":"25%","Backbone":"ResNet-50"}. This record does not assert independent reproduction.

### Taylor Pruning

Accuracy (%): 64.42. Result ID: 69b7c21047f34ce9249f7007.

[Reporting paper](<https://www.sota2.com/research/paper/69b78e930636e5bf4afbdc23.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65f11e209131cea0d448.md>)

Recorded evaluation configuration: {"Metric Proxy":"Loss Approx (∇W · W)","Capacity":"25%","Backbone":"ResNet-50"}. This record does not assert independent reproduction.

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