# Image Classification on DTD (Adversarial Accuracy and Latency)

Source: [Image Classification on DTD (Adversarial Accuracy and Latency)](<https://www.sota2.com/research/sota/image-classification-on-dtd-adversarial-accuracy-and-latency>)

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

Data updated: 2026-06-08T07:12:17.301Z

Evaluation of Image Classification performance on the Describable Textures Dataset (DTD), reporting adversarial robustness accuracy and average test time per image.

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) | DTD |
| Metric | Average Test Time (s) |
| Unit | TIME |
| Direction | Lower 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

- [Average Test Time (s)](<https://www.sota2.com/research/sota/image-classification-on-dtd-adversarial-accuracy-and-latency.md?metric=Average%20Test%20Time%20(s)>): Lower is better; unit: TIME.

- [Adversarial Accuracy (Robustness)](<https://www.sota2.com/research/sota/image-classification-on-dtd-adversarial-accuracy-and-latency.md?metric=Adversarial%20Accuracy%20(Robustness)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Average Test Time (s) | Result date | Report paper |
| --- | --- | --- | --- |
| CLIP | 0.4 | Not recorded | 6a264c5790966ade919be7e0 |
| TTC | 0.51 | Not recorded | 6a264c5790966ade919be7e0 |
| DOC | 0.56 | Not recorded | 6a264c5790966ade919be7e0 |
| SS-TPT | 0.75 | Not recorded | 6a264c5790966ade919be7e0 |
| Ensemble | 1.37 | Not recorded | 6a264c5790966ade919be7e0 |
| TPT | 1.44 | Not recorded | 6a264c5790966ade919be7e0 |
| R-TPT | 1.58 | Not recorded | 6a264c5790966ade919be7e0 |

### CLIP

Average Test Time (s): 0.4. Result ID: 6a266b10fbc373592c22b340.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5790966ade919be7e0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69581e209131cea21ad0.md>)

[Implementation](<https://github.com/OpenAI/CLIP>)

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

### TTC

Average Test Time (s): 0.51. Result ID: 6a266b10fbc373592c22b343.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5790966ade919be7e0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa774a1e209131cea724cc.md>)

[Implementation](<https://www.sota2.com/here>)

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

### DOC

Average Test Time (s): 0.56. Result ID: 6a266b10fbc373592c22b344.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5790966ade919be7e0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/691565ea0052cac1dc65d709.md>)

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

### SS-TPT

Average Test Time (s): 0.75. Result ID: 6a266b10fbc373592c22b346.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5790966ade919be7e0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a264c5790966ade919be7e0.md>)

Recorded evaluation configuration: {"# Aug. Views":"15"}. This record does not assert independent reproduction.

### Ensemble

Average Test Time (s): 1.37. Result ID: 6a266b10fbc373592c22b342.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5790966ade919be7e0.md>)

Recorded evaluation configuration: {"# Aug. Views":"63"}. This record does not assert independent reproduction.

### TPT

Average Test Time (s): 1.44. Result ID: 6a266b10fbc373592c22b341.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5790966ade919be7e0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6d691e209131cea399f9.md>)

[Implementation](<https://tinyurl.com/yr3zmhma>)

Recorded evaluation configuration: {"# Aug. Views":"63"}. This record does not assert independent reproduction.

### R-TPT

Average Test Time (s): 1.58. Result ID: 6a266b10fbc373592c22b345.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5790966ade919be7e0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7add1e209131cea8590d.md>)

[Implementation](<https://github.com/TomSheng21/R-TPT>)

Recorded evaluation configuration: {"# Aug. Views":"63"}. This record does not assert independent reproduction.

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