pyrit.prompt_converter.CharSwapConverter#
- class CharSwapConverter(*, max_iterations: int = 10, word_selection_strategy: WordSelectionStrategy | None = None)[source]#
Bases:
WordLevelConverterApplies character swapping to words in the prompt to test adversarial textual robustness.
- __init__(*, max_iterations: int = 10, word_selection_strategy: WordSelectionStrategy | None = None)[source]#
Initializes the converter with the specified parameters.
By default, 20% of randomly selected words will be perturbed.
- 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_selection_strategy (Optional[WordSelectionStrategy]) – Strategy for selecting which words to convert. If None, defaults to WordProportionSelectionStrategy(proportion=0.2).
Methods
__init__(*[, max_iterations, ...])Initializes the converter with the specified parameters.
convert_async(*, prompt[, input_type])Converts the given prompt into the target format supported by the converter.
convert_tokens_async(*, prompt[, ...])Converts substrings within a prompt that are enclosed by specified start and end tokens.
convert_word_async(word)Converts a single word into the target format supported by the converter.
get_identifier()Returns an identifier dictionary for the converter.
input_supported(input_type)Checks if the input type is supported by the converter.
join_words(words)Provides a way for subclasses to override the default behavior of joining words.
output_supported(output_type)Checks if the output type is supported by the converter.
validate_input(prompt)Validates the input before processing (can be overridden by subclasses).
Attributes
supported_input_typesReturns a list of supported input types for the converter.
supported_output_typesReturns a list of supported output types for the converter.