Summary of On Adversarial Examples For Text Classification by Perturbing Latent Representations, By Korn Sooksatra et al.
On Adversarial Examples for Text Classification by Perturbing Latent Representations
by Korn Sooksatra, Bikram Khanal, Pablo Rivas
First submitted to arxiv on: 6 May 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Cryptography and Security (cs.CR)
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 The paper presents a novel approach to assess the robustness of deep learning-based text classifiers against state-of-the-art attacks. It leverages the fact that text inputs are discrete and can be manipulated to create adversarial examples. To overcome previous limitations, the authors transform input texts into their embedding vectors, allowing for more effective white-box attacks. The proposed framework calculates the robustness of a classifier using its gradients. This innovative approach has important implications for developing more resilient text classification models. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper helps make text classifiers better by making them more resistant to bad data. Right now, these classifiers are vulnerable to fake inputs that can trick them into misclassifying texts. To fix this, the authors turn text inputs into vectors with real numbers instead of just 0s and 1s. This makes it easier to find clever ways to manipulate the input and make the classifier produce wrong results. The new approach also helps measure how robust a classifier is by looking at its gradients. |
Keywords
» Artificial intelligence » Deep learning » Embedding » Text classification
