no code implementations • 4 Nov 2024 • Atoosa Chegini, Hamid Kazemi, Iman Mirzadeh, Dong Yin, Maxwell Horton, Moin Nabi, Mehrdad Farajtabar, Keivan Alizadeh
As a result, policy optimization is often trapped in a narrow region of the parameter space, leading to suboptimal alignment and performance.
1 code implementation • 5 Mar 2024 • Hamid Kazemi, Atoosa Chegini, Jonas Geiping, Soheil Feizi, Tom Goldstein
We employ an inversion-based approach to examine CLIP models.
1 code implementation • 23 Feb 2024 • Vinu Sankar Sadasivan, Shoumik Saha, Gaurang Sriramanan, Priyatham Kattakinda, Atoosa Chegini, Soheil Feizi
Through human evaluations, we find that our untargeted attack causes Vicuna-7B-v1. 5 to produce ~15% more incorrect outputs when compared to LM outputs in the absence of our attack.
no code implementations • 9 Dec 2023 • Atoosa Chegini, Soheil Feizi
One common reason for these failures is the occurrence of objects in backgrounds that are rarely seen during training.
1 code implementation • 29 Sep 2023 • Mehrdad Saberi, Vinu Sankar Sadasivan, Keivan Rezaei, Aounon Kumar, Atoosa Chegini, Wenxiao Wang, Soheil Feizi
Moreover, we show that watermarking methods are vulnerable to spoofing attacks where the attacker aims to have real images identified as watermarked ones, damaging the reputation of the developers.
2 code implementations • 5 Feb 2023 • Keivan Rezaei, Kiarash Banihashem, Atoosa Chegini, Soheil Feizi
Based on this approach, we propose DPA+ROE and FA+ROE defense methods based on Deep Partition Aggregation (DPA) and Finite Aggregation (FA) approaches from prior work.