Picture for Taesung Lee

Taesung Lee

Towards Generating Informative Textual Description for Neurons in Language Models

Add code
Jan 30, 2024
Viaarxiv icon

URET: Universal Robustness Evaluation Toolkit (for Evasion)

Add code
Aug 03, 2023
Viaarxiv icon

Matching Pairs: Attributing Fine-Tuned Models to their Pre-Trained Large Language Models

Add code
Jun 15, 2023
Viaarxiv icon

Robustness of Explanation Methods for NLP Models

Add code
Jun 24, 2022
Figure 1 for Robustness of Explanation Methods for NLP Models
Figure 2 for Robustness of Explanation Methods for NLP Models
Figure 3 for Robustness of Explanation Methods for NLP Models
Figure 4 for Robustness of Explanation Methods for NLP Models
Viaarxiv icon

Adaptive Verifiable Training Using Pairwise Class Similarity

Add code
Dec 14, 2020
Figure 1 for Adaptive Verifiable Training Using Pairwise Class Similarity
Figure 2 for Adaptive Verifiable Training Using Pairwise Class Similarity
Figure 3 for Adaptive Verifiable Training Using Pairwise Class Similarity
Figure 4 for Adaptive Verifiable Training Using Pairwise Class Similarity
Viaarxiv icon

A new measure for overfitting and its implications for backdooring of deep learning

Add code
Jun 18, 2020
Figure 1 for A new measure for overfitting and its implications for backdooring of deep learning
Figure 2 for A new measure for overfitting and its implications for backdooring of deep learning
Figure 3 for A new measure for overfitting and its implications for backdooring of deep learning
Figure 4 for A new measure for overfitting and its implications for backdooring of deep learning
Viaarxiv icon

Detecting Backdoor Attacks on Deep Neural Networks by Activation Clustering

Add code
Nov 09, 2018
Figure 1 for Detecting Backdoor Attacks on Deep Neural Networks by Activation Clustering
Figure 2 for Detecting Backdoor Attacks on Deep Neural Networks by Activation Clustering
Figure 3 for Detecting Backdoor Attacks on Deep Neural Networks by Activation Clustering
Figure 4 for Detecting Backdoor Attacks on Deep Neural Networks by Activation Clustering
Viaarxiv icon

Defending Against Model Stealing Attacks Using Deceptive Perturbations

Add code
Sep 19, 2018
Figure 1 for Defending Against Model Stealing Attacks Using Deceptive Perturbations
Figure 2 for Defending Against Model Stealing Attacks Using Deceptive Perturbations
Figure 3 for Defending Against Model Stealing Attacks Using Deceptive Perturbations
Figure 4 for Defending Against Model Stealing Attacks Using Deceptive Perturbations
Viaarxiv icon