pyrit.prompt_converter.CharSwapGenerator

pyrit.prompt_converter.CharSwapGenerator#

class CharSwapGenerator(*, max_iterations: int = 10, word_swap_ratio: float = 0.2)[source]#

A PromptConverter that applies character swapping to words in the prompt to test adversarial textual robustness.

__init__(*, max_iterations: int = 10, word_swap_ratio: float = 0.2)[source]#

Initializes the CharSwapConverter.

Parameters:

max_iterations (int) – Number of times to generate perturbed prompts. The higher the number the higher the chance that words are different from the original prompt.
word_swap_ratio (float) – Percentage of words to perturb in the prompt per iteration.

Methods

`__init__`(*[, max_iterations, word_swap_ratio])	Initializes the CharSwapConverter.
`convert_async`(*, prompt[, input_type])	Converts the given prompt by applying character swaps.
`convert_tokens_async`(*, prompt[, ...])	Converts substrings within a prompt that are enclosed by specified start and end tokens.
`get_identifier`()
`input_supported`(input_type)	Checks if the input type is supported by the converter
`output_supported`(output_type)	Checks if the output type is supported by the converter

Attributes

`supported_input_types`	Returns a list of supported input types for the converter.
`supported_output_types`	Returns a list of supported output types for the converter.

async convert_async(*, prompt: str, input_type='text') → ConverterResult[source]#

Converts the given prompt by applying character swaps.

input_supported(input_type: Literal['text', 'image_path', 'audio_path', 'video_path', 'url', 'error']) → bool[source]#

Checks if the input type is supported by the converter

output_supported(output_type: Literal['text', 'image_path', 'audio_path', 'video_path', 'url', 'error']) → bool[source]#

Checks if the output type is supported by the converter