Summary of Exploiting Class Probabilities For Black-box Sentence-level Attacks, by Raha Moraffah and Huan Liu
Exploiting Class Probabilities for Black-box Sentence-level Attacks
by Raha Moraffah, Huan Liu
First submitted to arxiv on: 5 Feb 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR); Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary A novel algorithm for crafting adversarial sentences that exploit class probabilities in text classifiers has been developed, enabling more effective black-box sentence-level attacks. The proposed approach leverages class probability feedback to create misclassified sentences that are semantically equivalent to correctly classified ones. This research investigates the utility of class probabilities in enhancing attack success and evaluates the effectiveness of the novel algorithm across various classifiers and benchmark datasets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary A team of researchers has come up with a new way to trick text classification models by creating fake sentences that sound similar to real ones, but are actually wrong. They did this by using information from the model’s predictions to make the attacks stronger. This new method can be used even when we don’t know how the model is working internally. The scientists tested their approach on different models and datasets to see how well it worked. |
Keywords
* Artificial intelligence * Probability * Text classification
