promptcraft 0.4.0

PromptCraft: A Prompt Perturbation Toolkit for Prompt Robustness Analysis

PromptCraft

A Prompt Perturbation Toolkit for Prompt Robustness Analysis

Table of Contents

• Installation
• Character Editing
  • Character Replacement
  • Character Deletion
  • Character Insertion
  • Character Swap
  • Keyboard Typos
  • Optical Character Recognition (OCR)
• Word Manipulation
  • Synonym Replacement
  • Word Insertion
  • Word Swap
  • Word Deletion
  • Insert Punctuation
  • Word Split
• Sentence Paraphrasing
  • Back Translation based on 🤗 Hugging Face MarianMTModel
  • Back Translation based on Google Translator
  • Paraphrasing
  • Formal Style
  • Casual Style
  • Passive Style
  • Active Style
• Parallel Processing
• Structure of the Code
• Acknowledgement

Installation

pip install promptcraft

Character Editing

Character-level Prompt Perturbation
CharacterPerturb class for manipulating character in a sentence

from promptcraft import character

sentence = "Natalia sold clips to 48 of her friends in April, and then she sold half as many clips in May."
level = 0.25  # Percentage of characters that will be edited
character_tool = character.CharacterPerturb(sentence=sentence, level=level)

Character Replacement

Randomly replace level percentage characters from the sentence

char_replace = character_tool.character_replacement()

Character Deletion

Randomly delete level percentage characters from the sentence

char_delete = character_tool.character_deletion()

Character Insertion

Randomly insert level percentage characters to the sentence

char_insert = character_tool.character_insertion()

Character Swap

Randomly swap level percentage characters in the sentence
NOTE: including self-swapping

char_swap = character_tool.character_swap()

Keyboard Typos

Randomly substitute level percentage characters in the sentence with a randomly chosen character which is near the original character in the Keyboard (USA Full-size Layout)
NOTE:
(1) We applied keyboard_distance=1, i.e., the nearest character, number, or samples.
(2) If it is a character, we randomly chose lowercase or uppercase.

char_keyboard = character_tool.keyboard_typos()

Optical Character Recognition

Randomly substitute level percentage characters in the sentence with a common OCR map error

char_ocr = character_tool.optical_character_recognition()

Word Manipulation

Word-level Prompt Perturbation WordPerturb class for manipulating words in a sentence

NOTE: the number of words in a sentence is only the valid words without considering spaces, special symbols, and punctuations

from promptcraft import word

sentence = "Natalia sold clips to 48 of her friends in April, and then she sold half as many clips in May."
level = 0.25  # Percentage of words that will be manipulated
word_tool = word.WordPerturb(sentence=sentence, level=level)

Synonym Replacement

Randomly choose $n$ words from the sentence that are not stop words.
Replace each of these words with one of its synonyms chosen at random.
Problem 1: Without any synonyms
Problem 2: Fewer positions than needed positions

word_synonym = word_tool.synonym_replacement()

Word Insertion

Find a random synonym of a random word in the sentence that is not a stop word.
Insert that synonym into a random position in the sentence.
Do this $n$ times.

word_insert = word_tool.word_insertion()

Word Swap

Randomly choose two words in the sentence and swap their positions.
Do this $n$ times.

word_swap = word_tool.word_swap()

Word Deletion

Each word in the sentence can be randomly removed with probability $p$.

word_delete = word_tool.word_deletion()

Insert Punctuation

Randomly insert punctuation in the sentence with probability $p$.

word_punctuation = word_tool.insert_punctuation()

Word Split

Randomly split a word to two tokens randomly

word_split = word_tool.word_split()

Sentence Paraphrasing

Sentence-level Prompt Perturbation
SentencePerturb class for directly manipulating a sentence

from promptcraft import sentence

sen = "Natalia sold clips to 48 of her friends in April, and then she sold half as many clips in May."
sentence_tool = sentence.SentencePerturb(sentence=sen)

Back Translation by Hugging Face

Back translate the sentence (English $\rightarrow$ German $\rightarrow$ English) via 🤗 Hugging Face MarianMTModel

back_trans_hf = sentence_tool.back_translation_hugging_face()

Back Translation by Google Translator

Back translate the sentence (English $\rightarrow$ German $\rightarrow$ English) via Google Translate API

back_trans_google = sentence_tool.back_translation_google()

Paraphrasing

Paraphrasing the sentence via Parrot Paraphraser considering
(1) Adequency: Is the meaning preserved adequately?
(2) Fluency: Is the paraphrase fluent English?
(3) Diversity: (Lexical / Phrasal / Syntactical): How much has the paraphrase changed the original sentence?

sen_paraphrase = sentence_tool.paraphrase()

Formal Style

Transform the sentence style to Formal

sen_formal = sentence_tool.formal()

Casual Style

Transform the sentence style to Casual

sen_casual = sentence_tool.casual()

Passive Style

Transform the sentence style to Passive

sen_passive = sentence_tool.passive()

Active Style

Transform the sentence style to Active

sen_active = sentence_tool.active()

Parallel Processing

Since all the methods are executed on the CPU, they can be performed in parallel using the multiprocessing package.

Structure of the Code

At the root of the project, you will see:

.
├── LICENSE
├── README.md
├── promptcraft
│   ├── __init__.py
│   ├── character.py
│   ├── parrot.py
│   ├── sentence.py
│   ├── styleformer.py
│   └── word.py
└── setup.py

Acknowledgement

This work was finished during my 2023 fall semester research rotation at the Dependable Computing Laboratory, Department of Electrical and Computer Engineering, Boston University, under the supervision of Prof. Wenchao Li.