# Backdoor Detection on MS-COCO BadT2ISent v1.4 (test)

Source: [Backdoor Detection on MS-COCO BadT2ISent v1.4 (test)](<https://www.sota2.com/research/sota/backdoor-detection-on-ms-coco-badt2isent-v1-4-test>)

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

Data updated: 2026-02-26T05:31:16.494Z

Evaluation of backdoor detection performance on the MS-COCO BadT2ISent v1.4 test set, reporting Accuracy (Acc) and AUROC.

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 | Backdoor Detection |
| Dataset (including any stated split) | MS-COCO BadT2ISent v1.4 (test) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| 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/backdoor-detection-on-ms-coco-badt2isent-v1-4-test.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [AUROC](<https://www.sota2.com/research/sota/backdoor-detection-on-ms-coco-badt2isent-v1-4-test.md?metric=AUROC>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| TNC-Detect | 91.2 | Not recorded | 698181eea8b01e76cc8e295f |
| NaviT2I | 74.5 | Not recorded | 698181eea8b01e76cc8e295f |
| UFID | 58.9 | Not recorded | 698181eea8b01e76cc8e295f |
| T2IShield | 51.1 | Not recorded | 698181eea8b01e76cc8e295f |
| T2IShield | 49.6 | Not recorded | 698181eea8b01e76cc8e295f |

### TNC-Detect

Accuracy: 91.2. Result ID: 6996e1225698665ec42d9b09.

[Reporting paper](<https://www.sota2.com/research/paper/698181eea8b01e76cc8e295f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/698181eea8b01e76cc8e295f.md>)

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

### NaviT2I

Accuracy: 74.5. Result ID: 6996e1225698665ec42d9b08.

[Reporting paper](<https://www.sota2.com/research/paper/698181eea8b01e76cc8e295f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a651e209131cea82e92.md>)

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

### UFID

Accuracy: 58.9. Result ID: 6996e1225698665ec42d9b07.

[Reporting paper](<https://www.sota2.com/research/paper/698181eea8b01e76cc8e295f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76d41e209131cea6f613.md>)

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

### T2IShield

Accuracy: 51.1. Result ID: 6996e1225698665ec42d9b06.

[Reporting paper](<https://www.sota2.com/research/paper/698181eea8b01e76cc8e295f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74221e209131cea5fdbc.md>)

Recorded evaluation configuration: {"variant":"CDA"}. This record does not assert independent reproduction.

### T2IShield

Accuracy: 49.6. Result ID: 6996e1225698665ec42d9b05.

[Reporting paper](<https://www.sota2.com/research/paper/698181eea8b01e76cc8e295f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74221e209131cea5fdbc.md>)

Recorded evaluation configuration: {"variant":"FTT"}. This record does not assert independent reproduction.

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