Alert button
Picture for Prateek Mittal

Prateek Mittal

Alert button

Assessing the Brittleness of Safety Alignment via Pruning and Low-Rank Modifications

Feb 07, 2024
Boyi Wei, Kaixuan Huang, Yangsibo Huang, Tinghao Xie, Xiangyu Qi, Mengzhou Xia, Prateek Mittal, Mengdi Wang, Peter Henderson

Viaarxiv icon

Efficient Data Shapley for Weighted Nearest Neighbor Algorithms

Jan 20, 2024
Jiachen T. Wang, Prateek Mittal, Ruoxi Jia

Viaarxiv icon

Private Fine-tuning of Large Language Models with Zeroth-order Optimization

Jan 09, 2024
Xinyu Tang, Ashwinee Panda, Milad Nasr, Saeed Mahloujifar, Prateek Mittal

Viaarxiv icon

PatchCURE: Improving Certifiable Robustness, Model Utility, and Computation Efficiency of Adversarial Patch Defenses

Oct 19, 2023
Chong Xiang, Tong Wu, Sihui Dai, Jonathan Petit, Suman Jana, Prateek Mittal

Viaarxiv icon

Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!

Oct 05, 2023
Xiangyu Qi, Yi Zeng, Tinghao Xie, Pin-Yu Chen, Ruoxi Jia, Prateek Mittal, Peter Henderson

Figure 1 for Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!
Figure 2 for Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!
Figure 3 for Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!
Figure 4 for Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!
Viaarxiv icon

Threshold KNN-Shapley: A Linear-Time and Privacy-Friendly Approach to Data Valuation

Aug 30, 2023
Jiachen T. Wang, Yuqing Zhu, Yu-Xiang Wang, Ruoxi Jia, Prateek Mittal

Figure 1 for Threshold KNN-Shapley: A Linear-Time and Privacy-Friendly Approach to Data Valuation
Figure 2 for Threshold KNN-Shapley: A Linear-Time and Privacy-Friendly Approach to Data Valuation
Figure 3 for Threshold KNN-Shapley: A Linear-Time and Privacy-Friendly Approach to Data Valuation
Figure 4 for Threshold KNN-Shapley: A Linear-Time and Privacy-Friendly Approach to Data Valuation
Viaarxiv icon

BaDExpert: Extracting Backdoor Functionality for Accurate Backdoor Input Detection

Aug 23, 2023
Tinghao Xie, Xiangyu Qi, Ping He, Yiming Li, Jiachen T. Wang, Prateek Mittal

Figure 1 for BaDExpert: Extracting Backdoor Functionality for Accurate Backdoor Input Detection
Figure 2 for BaDExpert: Extracting Backdoor Functionality for Accurate Backdoor Input Detection
Figure 3 for BaDExpert: Extracting Backdoor Functionality for Accurate Backdoor Input Detection
Figure 4 for BaDExpert: Extracting Backdoor Functionality for Accurate Backdoor Input Detection
Viaarxiv icon

Food Classification using Joint Representation of Visual and Textual Data

Aug 03, 2023
Prateek Mittal, Puneet Goyal, Joohi Chauhan

Viaarxiv icon

Visual Adversarial Examples Jailbreak Large Language Models

Jun 22, 2023
Xiangyu Qi, Kaixuan Huang, Ashwinee Panda, Mengdi Wang, Prateek Mittal

Figure 1 for Visual Adversarial Examples Jailbreak Large Language Models
Figure 2 for Visual Adversarial Examples Jailbreak Large Language Models
Figure 3 for Visual Adversarial Examples Jailbreak Large Language Models
Figure 4 for Visual Adversarial Examples Jailbreak Large Language Models
Viaarxiv icon

Differentially Private Image Classification by Learning Priors from Random Processes

Jun 08, 2023
Xinyu Tang, Ashwinee Panda, Vikash Sehwag, Prateek Mittal

Figure 1 for Differentially Private Image Classification by Learning Priors from Random Processes
Figure 2 for Differentially Private Image Classification by Learning Priors from Random Processes
Figure 3 for Differentially Private Image Classification by Learning Priors from Random Processes
Figure 4 for Differentially Private Image Classification by Learning Priors from Random Processes
Viaarxiv icon