Add pipeline tag and link to paper (#1)
Browse files- Add pipeline tag and link to paper (b6835e8158e6d5c8848cef7e9e7d7e85fdde31ee)
Co-authored-by: Niels Rogge <nielsr@users.noreply.huggingface.co>
README.md
CHANGED
|
@@ -1,18 +1,21 @@
|
|
| 1 |
---
|
| 2 |
-
|
|
|
|
| 3 |
language:
|
| 4 |
- en
|
|
|
|
|
|
|
| 5 |
tags:
|
| 6 |
- prompt-injection
|
| 7 |
- indirect-prompt-injection
|
| 8 |
- agent-security
|
| 9 |
- agentdojo
|
| 10 |
-
datasets:
|
| 11 |
-
- tavialiu/CAD-data
|
| 12 |
---
|
| 13 |
|
| 14 |
# CAD Models
|
| 15 |
|
|
|
|
|
|
|
| 16 |
Classifier checkpoints for CAD, from *Robust Context-Aware Detection of Malicious Instructions
|
| 17 |
in Text* (Liu, Ma, and Vorobeychik, 2026).
|
| 18 |
|
|
@@ -57,4 +60,4 @@ bash scripts/run_agentdojo_benchmark.sh banking slack travel
|
|
| 57 |
journal = {arXiv preprint},
|
| 58 |
year = {2026}
|
| 59 |
}
|
| 60 |
-
```
|
|
|
|
| 1 |
---
|
| 2 |
+
datasets:
|
| 3 |
+
- tavialiu/CAD-data
|
| 4 |
language:
|
| 5 |
- en
|
| 6 |
+
license: mit
|
| 7 |
+
pipeline_tag: text-classification
|
| 8 |
tags:
|
| 9 |
- prompt-injection
|
| 10 |
- indirect-prompt-injection
|
| 11 |
- agent-security
|
| 12 |
- agentdojo
|
|
|
|
|
|
|
| 13 |
---
|
| 14 |
|
| 15 |
# CAD Models
|
| 16 |
|
| 17 |
+
Paper: [Robust Context-Aware Detection of Malicious Instructions in Text](https://huggingface.co/papers/2608.05430)
|
| 18 |
+
|
| 19 |
Classifier checkpoints for CAD, from *Robust Context-Aware Detection of Malicious Instructions
|
| 20 |
in Text* (Liu, Ma, and Vorobeychik, 2026).
|
| 21 |
|
|
|
|
| 60 |
journal = {arXiv preprint},
|
| 61 |
year = {2026}
|
| 62 |
}
|
| 63 |
+
```
|