no code implementations • 16 Feb 2024 • Hae Jin Song, Mahyar Khayatkhoei, Wael AbdAlmageed
Recent works have shown that generative models leave traces of their underlying generative process on the generated samples, broadly referred to as fingerprints of a generative model, and have studied their utility in detecting synthetic images from real ones.
1 code implementation • 12 Feb 2024 • Jiarui Zhang, Jinyi Hu, Mahyar Khayatkhoei, Filip Ilievski, Maosong Sun
Multimodal Large Language Models (MLLMs) have recently shown remarkable perceptual capability in answering visual questions, however, little is known about the limits of their perception.
no code implementations • 28 Nov 2023 • Mulin Tian, Mahyar Khayatkhoei, Joe Mathai, Wael AbdAlmageed
Deepfake videos present an increasing threat to society with potentially negative impact on criminal justice, democracy, and personal safety and privacy.
1 code implementation • 13 Nov 2023 • Jiazhi Li, Mahyar Khayatkhoei, Jiageng Zhu, Hanchen Xie, Mohamed E. Hussein, Wael AbdAlmageed
To that end, in this work, we mathematically and empirically reveal the limitation of existing attribute bias removal methods in presence of strong bias and propose a new method that can mitigate this limitation.
2 code implementations • 24 Oct 2023 • Jiarui Zhang, Mahyar Khayatkhoei, Prateek Chhikara, Filip Ilievski
In particular, we show that their zero-shot accuracy in answering visual questions is very sensitive to the size of the visual subject of the question, declining up to 46% with size.
1 code implementation • 8 Oct 2023 • Jiazhi Li, Mahyar Khayatkhoei, Jiageng Zhu, Hanchen Xie, Mohamed E. Hussein, Wael AbdAlmageed
Ensuring a neural network is not relying on protected attributes (e. g., race, sex, age) for predictions is crucial in advancing fair and trustworthy AI.
1 code implementation • 10 Aug 2023 • Jiageng Zhu, Hanchen Xie, Jianhua Wu, Jiazhi Li, Mahyar Khayatkhoei, Mohamed E. Hussein, Wael AbdAlmageed
Most causal representation learning (CRL) methods are fully supervised, which is impractical due to costly labeling.
1 code implementation • 16 Jun 2023 • Mahyar Khayatkhoei, Wael AbdAlmageed
Precision and Recall are two prominent metrics of generative performance, which were proposed to separately measure the fidelity and diversity of generative models.
no code implementations • 31 May 2023 • Jiarui Zhang, Mahyar Khayatkhoei, Prateek Chhikara, Filip Ilievski
As our initial analysis of BLIP-family models revealed difficulty with answering fine-detail questions, we investigate the following question: Can visual cropping be employed to improve the performance of state-of-the-art visual question answering models on fine-detail questions?
1 code implementation • 12 May 2023 • Hanchen Xie, Jiageng Zhu, Mahyar Khayatkhoei, Jiazhi Li, Mohamed E. Hussein, Wael AbdAlmageed
In this paper, we investigate two challenging conditions for environment misalignment: Cross-Domain and Cross-Context by proposing four datasets that are designed for these challenges: SimB-Border, SimB-Split, BlenB-Border, and BlenB-Split.
no code implementations • 4 Oct 2020 • Mahyar Khayatkhoei, Ahmed Elgammal
As the success of Generative Adversarial Networks (GANs) on natural images quickly propels them into various real-life applications across different domains, it becomes more and more important to clearly understand their limitations.
1 code implementation • NeurIPS 2018 • Mahyar Khayatkhoei, Ahmed Elgammal, Maneesh Singh
Natural images may lie on a union of disjoint manifolds rather than one globally connected manifold, and this can cause several difficulties for the training of common Generative Adversarial Networks (GANs).