Alert button
Picture for Sarah Erfani

Sarah Erfani

Alert button

Round Trip Translation Defence against Large Language Model Jailbreaking Attacks

Add code
Bookmark button
Alert button
Feb 21, 2024
Canaan Yung, Hadi Mohaghegh Dolatabadi, Sarah Erfani, Christopher Leckie

Figure 1 for Round Trip Translation Defence against Large Language Model Jailbreaking Attacks
Figure 2 for Round Trip Translation Defence against Large Language Model Jailbreaking Attacks
Figure 3 for Round Trip Translation Defence against Large Language Model Jailbreaking Attacks
Figure 4 for Round Trip Translation Defence against Large Language Model Jailbreaking Attacks
Viaarxiv icon

OIL-AD: An Anomaly Detection Framework for Sequential Decision Sequences

Add code
Bookmark button
Alert button
Feb 07, 2024
Chen Wang, Sarah Erfani, Tansu Alpcan, Christopher Leckie

Viaarxiv icon

The Devil's Advocate: Shattering the Illusion of Unexploitable Data using Diffusion Models

Add code
Bookmark button
Alert button
Mar 15, 2023
Hadi M. Dolatabadi, Sarah Erfani, Christopher Leckie

Figure 1 for The Devil's Advocate: Shattering the Illusion of Unexploitable Data using Diffusion Models
Figure 2 for The Devil's Advocate: Shattering the Illusion of Unexploitable Data using Diffusion Models
Figure 3 for The Devil's Advocate: Shattering the Illusion of Unexploitable Data using Diffusion Models
Figure 4 for The Devil's Advocate: Shattering the Illusion of Unexploitable Data using Diffusion Models
Viaarxiv icon

Distilling Cognitive Backdoor Patterns within an Image

Add code
Bookmark button
Alert button
Jan 26, 2023
Hanxun Huang, Xingjun Ma, Sarah Erfani, James Bailey

Figure 1 for Distilling Cognitive Backdoor Patterns within an Image
Figure 2 for Distilling Cognitive Backdoor Patterns within an Image
Figure 3 for Distilling Cognitive Backdoor Patterns within an Image
Figure 4 for Distilling Cognitive Backdoor Patterns within an Image
Viaarxiv icon

COLLIDER: A Robust Training Framework for Backdoor Data

Add code
Bookmark button
Alert button
Oct 13, 2022
Hadi M. Dolatabadi, Sarah Erfani, Christopher Leckie

Figure 1 for COLLIDER: A Robust Training Framework for Backdoor Data
Figure 2 for COLLIDER: A Robust Training Framework for Backdoor Data
Figure 3 for COLLIDER: A Robust Training Framework for Backdoor Data
Figure 4 for COLLIDER: A Robust Training Framework for Backdoor Data
Viaarxiv icon

Adversarial Coreset Selection for Efficient Robust Training

Add code
Bookmark button
Alert button
Sep 13, 2022
Hadi M. Dolatabadi, Sarah Erfani, Christopher Leckie

Figure 1 for Adversarial Coreset Selection for Efficient Robust Training
Figure 2 for Adversarial Coreset Selection for Efficient Robust Training
Figure 3 for Adversarial Coreset Selection for Efficient Robust Training
Figure 4 for Adversarial Coreset Selection for Efficient Robust Training
Viaarxiv icon

Detecting Arbitrary Order Beneficial Feature Interactions for Recommender Systems

Add code
Bookmark button
Alert button
Jun 28, 2022
Yixin Su, Yunxiang Zhao, Sarah Erfani, Junhao Gan, Rui Zhang

Figure 1 for Detecting Arbitrary Order Beneficial Feature Interactions for Recommender Systems
Figure 2 for Detecting Arbitrary Order Beneficial Feature Interactions for Recommender Systems
Figure 3 for Detecting Arbitrary Order Beneficial Feature Interactions for Recommender Systems
Figure 4 for Detecting Arbitrary Order Beneficial Feature Interactions for Recommender Systems
Viaarxiv icon

Robust Task-Oriented Dialogue Generation with Contrastive Pre-training and Adversarial Filtering

Add code
Bookmark button
Alert button
May 20, 2022
Shiquan Yang, Xinting Huang, Jey Han Lau, Sarah Erfani

Figure 1 for Robust Task-Oriented Dialogue Generation with Contrastive Pre-training and Adversarial Filtering
Figure 2 for Robust Task-Oriented Dialogue Generation with Contrastive Pre-training and Adversarial Filtering
Figure 3 for Robust Task-Oriented Dialogue Generation with Contrastive Pre-training and Adversarial Filtering
Figure 4 for Robust Task-Oriented Dialogue Generation with Contrastive Pre-training and Adversarial Filtering
Viaarxiv icon

An Interpretable Neuro-Symbolic Reasoning Framework for Task-Oriented Dialogue Generation

Add code
Bookmark button
Alert button
Mar 11, 2022
Shiquan Yang, Rui Zhang, Sarah Erfani, Jey Han Lau

Figure 1 for An Interpretable Neuro-Symbolic Reasoning Framework for Task-Oriented Dialogue Generation
Figure 2 for An Interpretable Neuro-Symbolic Reasoning Framework for Task-Oriented Dialogue Generation
Figure 3 for An Interpretable Neuro-Symbolic Reasoning Framework for Task-Oriented Dialogue Generation
Figure 4 for An Interpretable Neuro-Symbolic Reasoning Framework for Task-Oriented Dialogue Generation
Viaarxiv icon

$\ell_\infty$-Robustness and Beyond: Unleashing Efficient Adversarial Training

Add code
Bookmark button
Alert button
Dec 01, 2021
Hadi M. Dolatabadi, Sarah Erfani, Christopher Leckie

Figure 1 for $\ell_\infty$-Robustness and Beyond: Unleashing Efficient Adversarial Training
Figure 2 for $\ell_\infty$-Robustness and Beyond: Unleashing Efficient Adversarial Training
Figure 3 for $\ell_\infty$-Robustness and Beyond: Unleashing Efficient Adversarial Training
Figure 4 for $\ell_\infty$-Robustness and Beyond: Unleashing Efficient Adversarial Training
Viaarxiv icon