Zobrazeno 1 - 10
of 3 227
pro vyhledávání: '"Chegini A"'
We employ an inversion-based approach to examine CLIP models. Our examination reveals that inverting CLIP models results in the generation of images that exhibit semantic alignment with the specified target prompts. We leverage these inverted images
Externí odkaz:
http://arxiv.org/abs/2403.02580
Autor:
Sadasivan, Vinu Sankar, Saha, Shoumik, Sriramanan, Gaurang, Kattakinda, Priyatham, Chegini, Atoosa, Feizi, Soheil
In this paper, we introduce a novel class of fast, beam search-based adversarial attack (BEAST) for Language Models (LMs). BEAST employs interpretable parameters, enabling attackers to balance between attack speed, success rate, and the readability o
Externí odkaz:
http://arxiv.org/abs/2402.15570
Autor:
Chegini, Atoosa, Feizi, Soheil
Deep learning models can encounter unexpected failures, especially when dealing with challenging sub-populations. One common reason for these failures is the occurrence of objects in backgrounds that are rarely seen during training. To gain a better
Externí odkaz:
http://arxiv.org/abs/2312.05464
The locality of solution features in cardiac electrophysiology simulations calls for adaptive methods. Due to the overhead incurred by established mesh refinement and coarsening, however, such approaches failed in accelerating the computations. Here
Externí odkaz:
http://arxiv.org/abs/2311.07206
Autor:
Basu, Samyadeep, Saberi, Mehrdad, Bhardwaj, Shweta, Chegini, Atoosa Malemir, Massiceti, Daniela, Sanjabi, Maziar, Hu, Shell Xu, Feizi, Soheil
A plethora of text-guided image editing methods have recently been developed by leveraging the impressive capabilities of large-scale diffusion-based generative models such as Imagen and Stable Diffusion. A standardized evaluation protocol, however,
Externí odkaz:
http://arxiv.org/abs/2310.02426
Autor:
Saberi, Mehrdad, Sadasivan, Vinu Sankar, Rezaei, Keivan, Kumar, Aounon, Chegini, Atoosa, Wang, Wenxiao, Feizi, Soheil
In light of recent advancements in generative AI models, it has become essential to distinguish genuine content from AI-generated one to prevent the malicious usage of fake materials as authentic ones and vice versa. Various techniques have been intr
Externí odkaz:
http://arxiv.org/abs/2310.00076
Autor:
Raziye Chegini, Morteza Sadeghi, Sadegh Shirian, Fatemeh Sabbaghziarani, Ehsan Aali, Pouriya Soleimani, Mohammad Reza Ashtari Majelan, Fariba Zafari, Shahram Darabi
Publikováno v:
International Journal of Reproductive BioMedicine, Vol 22, Iss 7, Pp 527-538 (2024)
Abstract Background: Melatonin and L-carnitine are free radical scavengers with antiapoptotic and antioxidant properties that improve oocyte development. Objective: This study aimed to find the possible effect of combining 2 antioxidant agents of mel
Externí odkaz:
https://doaj.org/article/950e8d2f4eff4018bb452467d94c626d
Autor:
Chegini, Sadegh, Zarepour, Mahmoud
An important functional of Poisson random measure is the negative binomial process (NBP). We use NBP to introduce a generalized Poisson-Kingman distribution and its corresponding random discrete probability measure. This random discrete probability m
Externí odkaz:
http://arxiv.org/abs/2307.00176
Autor:
Lu, Jin, Li, Xingpeng, Li, Hongyi, Chegini, Taher, Gamarra, Carlos, Yang, Y. C. Ethan, Cook, Margaret, Dillingham, Gavin
This study introduced a synthetic power system with spatio-temporally correlated profiles of solar power, wind power, dynamic line ratings and loads at one-hour resolution for five continuous years, referred to as the Texas 123-bus backbone transmiss
Externí odkaz:
http://arxiv.org/abs/2302.13231
In data poisoning attacks, an adversary tries to change a model's prediction by adding, modifying, or removing samples in the training data. Recently, ensemble-based approaches for obtaining provable defenses against data poisoning have been proposed
Externí odkaz:
http://arxiv.org/abs/2302.02300