# Image Classification on Vision Task Mixture Unseen Tasks 4-task (test)

Source: [Image Classification on Vision Task Mixture Unseen Tasks 4-task (test)](<https://www.sota2.com/research/sota/image-classification-on-vision-task-mixture-unseen-tasks-4-task-test>)

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

Data updated: 2026-03-09T08:22:48.618Z

Evaluation of Image Classification performance across the four specified tasks (CIFAR100, Flowers, Pets, STL10) included in the Vision Task Mixture Unseen Tasks suite, reported on the test split.

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) | Vision Task Mixture Unseen Tasks 4-task (test) |
| Metric | Accuracy (CIFAR100) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 7 of 7 |
| 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 (CIFAR100)](<https://www.sota2.com/research/sota/image-classification-on-vision-task-mixture-unseen-tasks-4-task-test.md?metric=Accuracy%20(CIFAR100)>): Higher is better; unit: PERCENT.

- [Accuracy (Flowers)](<https://www.sota2.com/research/sota/image-classification-on-vision-task-mixture-unseen-tasks-4-task-test.md?metric=Accuracy%20(Flowers)>): Higher is better; unit: PERCENT.

- [Accuracy (Pets)](<https://www.sota2.com/research/sota/image-classification-on-vision-task-mixture-unseen-tasks-4-task-test.md?metric=Accuracy%20(Pets)>): Higher is better; unit: PERCENT.

- [Accuracy (STL10)](<https://www.sota2.com/research/sota/image-classification-on-vision-task-mixture-unseen-tasks-4-task-test.md?metric=Accuracy%20(STL10)>): Higher is better; unit: PERCENT.

- [Average Accuracy](<https://www.sota2.com/research/sota/image-classification-on-vision-task-mixture-unseen-tasks-4-task-test.md?metric=Average%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy (CIFAR100) | Result date | Report paper |
| --- | --- | --- | --- |
| Zeroshot | 64.61 | Not recorded | 69ae53e62d603e6de3e4294c |
| DC-Merge | 56.9 | Not recorded | 69ae53e62d603e6de3e4294c |
| Iso-CTS | 52.64 | Not recorded | 69ae53e62d603e6de3e4294c |
| TSV-M | 51.13 | Not recorded | 69ae53e62d603e6de3e4294c |
| KnOTS-TIES | 47.21 | Not recorded | 69ae53e62d603e6de3e4294c |
| WUDI-Merging | 46.81 | Not recorded | 69ae53e62d603e6de3e4294c |
| Task Arithmetic | 43.83 | Not recorded | 69ae53e62d603e6de3e4294c |

### Zeroshot

Accuracy (CIFAR100): 64.61. Result ID: 69ae83436dbb7075b9569a6c.

[Reporting paper](<https://www.sota2.com/research/paper/69ae53e62d603e6de3e4294c.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B-32","Fine-tuning setting":"LoRA"}. This record does not assert independent reproduction.

### DC-Merge

Accuracy (CIFAR100): 56.9. Result ID: 69ae83436dbb7075b9569a72.

[Reporting paper](<https://www.sota2.com/research/paper/69ae53e62d603e6de3e4294c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69ae53e62d603e6de3e4294c.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B-32","Fine-tuning setting":"LoRA"}. This record does not assert independent reproduction.

### Iso-CTS

Accuracy (CIFAR100): 52.64. Result ID: 69ae83436dbb7075b9569a71.

[Reporting paper](<https://www.sota2.com/research/paper/69ae53e62d603e6de3e4294c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa795d1e209131cea7d4a1.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B-32","Fine-tuning setting":"LoRA"}. This record does not assert independent reproduction.

### TSV-M

Accuracy (CIFAR100): 51.13. Result ID: 69ae83436dbb7075b9569a70.

[Reporting paper](<https://www.sota2.com/research/paper/69ae53e62d603e6de3e4294c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77c11e209131cea751d7.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B-32","Fine-tuning setting":"LoRA"}. This record does not assert independent reproduction.

### KnOTS-TIES

Accuracy (CIFAR100): 47.21. Result ID: 69ae83436dbb7075b9569a6e.

[Reporting paper](<https://www.sota2.com/research/paper/69ae53e62d603e6de3e4294c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa755b1e209131cea676d6.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B-32","Fine-tuning setting":"LoRA"}. This record does not assert independent reproduction.

### WUDI-Merging

Accuracy (CIFAR100): 46.81. Result ID: 69ae83436dbb7075b9569a6f.

[Reporting paper](<https://www.sota2.com/research/paper/69ae53e62d603e6de3e4294c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa795e1e209131cea7d4c8.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B-32","Fine-tuning setting":"LoRA"}. This record does not assert independent reproduction.

### Task Arithmetic

Accuracy (CIFAR100): 43.83. Result ID: 69ae83436dbb7075b9569a6d.

[Reporting paper](<https://www.sota2.com/research/paper/69ae53e62d603e6de3e4294c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ef01e209131cea43012.md>)

Recorded evaluation configuration: {"Backbone":"ViT-B-32","Fine-tuning setting":"LoRA"}. This record does not assert independent reproduction.

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