Azure
[Multiple Tasks] FEAT add DecodingTrust dataset
Is your feature request related to a problem? Please describe.
DecodingTrust data should be available via PyRIT https://github.com/AI-secure/DecodingTrust
Describe the solution you'd like
There should be a fetch function similar to #267 in pyrit.datasets. See completed PR for Stereotypes as an example.
Additional context
The datasets within the task list need a bit more evaluation to understand if they are useful. Datasets are relevant if they contain data that can be used for red teaming prompts. If you'd like to pick up one of the tasks, create an Issue from the task and assign yourself to it.
### Tasks
- [x] Stereotypes: https://github.com/Azure/PyRIT/pull/385
- [ ] Toxicity
- [ ] Adversarial Robustness
- [ ] OOD Robustness
- [ ] Robustness on Adversarial Demonstrations
- [ ] Privacy
- [ ] Machine Ethics
- [ ] Fairness
Added ability to fetch the 'Stereotypes' prompts data, but there are still 7 other trustworthiness perspectives with data that may (or some may not) be useful in PyRIT! Take a look here: https://github.com/AI-secure/DecodingTrust/tree/main/data
Take into consideration, before picking up a dataset, that the contents of these datasets include profanity and topics that you may not want to deal with.
