Seyed-Mohsen Moosavi-Dezfooli

66 papers A* 21A 1B 1Misc 1Journal 38Unranked 3

Year	Rank	Type	Title / Venue / Authors
2026	J	`jnl`	The Potential of CoT for Reasoning: A Closer Look at Trace Dynamics. CoRR Gregor Bachmann, Yichen Jiang, Seyed-Mohsen Moosavi-Dezfooli, Moin Nabi
2026	J	`jnl`	Thinking into the Future: Latent Lookahead Training for Transformers. CoRR Lorenzo Noci, Gregor Bachmann, Seyed-Mohsen Moosavi-Dezfooli, Moin Nabi
2025	A*	`conf`	Certified Human Trajectory Prediction. CVPR Mohammadhossein Bahari, Saeed Saadatnejad, Amirhossein Askari-Farsangi, Seyed-Mohsen Moosavi-Dezfooli, Alexandre Alahi
2025	A*	`conf`	Geometric Inductive Biases of Deep Networks: The Role of Data and Architecture. ICLR Sajad Movahedi, Antonio Orvieto, Seyed-Mohsen Moosavi-Dezfooli
2025	J	`jnl`	LORE: Lagrangian-Optimized Robust Embeddings for Visual Encoders. CoRR Borna Khodabandeh, Amirabbas Afzali, Amirhossein Afsharrad, Seyed Shahabeddin Mousavi, Sanjay Lall, Sajjad Amini, Seyed-Mohsen Moosavi-Dezfooli
2025	J	`jnl`	On the Anisotropy of Score-Based Generative Models. CoRR Andreas Floros, Seyed-Mohsen Moosavi-Dezfooli, Pier Luigi Dragotti
2025	J	`jnl`	Rewriting the Budget: A General Framework for Black-Box Attacks Under Cost Asymmetry. CoRR Mahdi Salmani, Alireza Abdollahpoorrostam, Seyed-Mohsen Moosavi-Dezfooli
2025	J	`jnl`	Trustworthy Image Super-Resolution via Generative Pseudoinverse. CoRR Andreas Floros, Seyed-Mohsen Moosavi-Dezfooli, Pier Luigi Dragotti
2024	J	`jnl`	Certified Human Trajectory Prediction. CoRR Mohammadhossein Bahari, Saeed Saadatnejad, Amirhossein Asgari-Farsangi, Seyed-Mohsen Moosavi-Dezfooli, Alexandre Alahi
2024	J	`jnl`	Geometric Inductive Biases of Deep Networks: The Role of Data and Architecture. CoRR Sajad Movahedi, Antonio Orvieto, Seyed-Mohsen Moosavi-Dezfooli
2024	A*	`conf`	SuperDeepFool: a new fast and accurate minimal adversarial attack. NeurIPS Alireza Abdollahpour, Mahed Abroshan, Seyed-Mohsen Moosavi-Dezfooli
2024	J	`jnl`	Tracing the Roots: Leveraging Temporal Dynamics in Diffusion Trajectories for Origin Attribution. CoRR Andreas Floros, Seyed-Mohsen Moosavi-Dezfooli, Pier Luigi Dragotti
2023	A*	`conf`	How to choose your best allies for a transferable attack? ICCV Thibault Maho, Seyed-Mohsen Moosavi-Dezfooli, Teddy Furon
2023	J	`jnl`	How to choose your best allies for a transferable attack? CoRR Thibault Maho, Seyed-Mohsen Moosavi-Dezfooli, Teddy Furon
2023	J	`jnl`	Revisiting DeepFool: generalization and improvement. CoRR Alireza Abdollahpourrostam, Mahed Abroshan, Seyed-Mohsen Moosavi-Dezfooli
2023	A*	`conf`	Sparse Attacks for Manipulating Explanations in Deep Neural Network Models. ICDM Ahmad Ajalloeian, Seyed-Mohsen Moosavi-Dezfooli, Michalis Vlachos, Pascal Frossard
2023	A*	`conf`	The Enemy of My Enemy is My Friend: Exploring Inverse Adversaries for Improving Adversarial Training. CVPR Junhao Dong, Seyed-Mohsen Moosavi-Dezfooli, Jianhuang Lai, Xiaohua Xie
2022	A	`conf`	On Smoothed Explanations: Quality and Robustness. CIKM Ahmad Ajalloeian, Seyed-Mohsen Moosavi-Dezfooli, Michalis Vlachos, Pascal Frossard
2022	—	`conf`	PRIME: A Few Primitives Can Boost Robustness to Common Corruptions. ECCV (25) Apostolos Modas, Rahul Rade, Guillermo Ortiz-Jiménez, Seyed-Mohsen Moosavi-Dezfooli, Pascal Frossard
2022	A*	`conf`	Reducing Excessive Margin to Achieve a Better Accuracy vs. Robustness Trade-off. ICLR Rahul Rade, Seyed-Mohsen Moosavi-Dezfooli
2022	J	`jnl`	The Enemy of My Enemy is My Friend: Exploring Inverse Adversaries for Improving Adversarial Training. CoRR Junhao Dong, Seyed-Mohsen Moosavi-Dezfooli, Jianhuang Lai, Xiaohua Xie
2022	A*	`conf`	Vehicle trajectory prediction works, but not everywhere. CVPR Mohammadhossein Bahari, Saeed Saadatnejad, Ahmad Rahimi, Mohammad Shahverdikondori, Amir Hossein Shahidzadeh, Seyed-Mohsen Moosavi-Dezfooli, Alexandre Alahi
2021	J	`jnl`	A neural anisotropic view of underspecification in deep learning. CoRR Guillermo Ortiz-Jiménez, Itamar Franco Salazar-Reque, Apostolos Modas, Seyed-Mohsen Moosavi-Dezfooli, Pascal Frossard
2021	J	`jnl`	Adversarial training may be a double-edged sword. CoRR Ali Rahmati, Seyed-Mohsen Moosavi-Dezfooli, Huaiyu Dai
2021	J	`jnl`	Are socially-aware trajectory prediction models really socially-aware? CoRR Saeed Saadatnejad, Mohammadhossein Bahari, Pedram Khorsandi, Mohammad Saneian, Seyed-Mohsen Moosavi-Dezfooli, Alexandre Alahi
2021	J	`jnl`	Optimism in the Face of Adversity: Understanding and Improving Deep Learning Through Adversarial Robustness. Proc. IEEE Guillermo Ortiz-Jiménez, Apostolos Modas, Seyed-Mohsen Moosavi-Dezfooli, Pascal Frossard
2021	J	`jnl`	PRIME: A Few Primitives Can Boost Robustness to Common Corruptions. CoRR Apostolos Modas, Rahul Rade, Guillermo Ortiz-Jiménez, Seyed-Mohsen Moosavi-Dezfooli, Pascal Frossard
2021	J	`jnl`	Understanding Catastrophic Overfitting in Adversarial Training. CoRR Peilin Kang, Seyed-Mohsen Moosavi-Dezfooli
2021	A*	`conf`	Uniform Convergence, Adversarial Spheres and a Simple Remedy. ICML Gregor Bachmann, Seyed-Mohsen Moosavi-Dezfooli, Thomas Hofmann
2021	J	`jnl`	Uniform Convergence, Adversarial Spheres and a Simple Remedy. CoRR Gregor Bachmann, Seyed-Mohsen Moosavi-Dezfooli, Thomas Hofmann
2021	J	`jnl`	Vehicle trajectory prediction works, but not everywhere. CoRR Mohammadhossein Bahari, Saeed Saadatnejad, Ahmad Rahimi, Mohammad Shahverdikondori, Mohammad Shahidzadeh, Seyed-Mohsen Moosavi-Dezfooli, Alexandre Alahi
2021	A*	`conf`	What can linearized neural networks actually say about generalization? NeurIPS Guillermo Ortiz-Jiménez, Seyed-Mohsen Moosavi-Dezfooli, Pascal Frossard
2021	J	`jnl`	What can linearized neural networks actually say about generalization? CoRR Guillermo Ortiz-Jiménez, Seyed-Mohsen Moosavi-Dezfooli, Pascal Frossard
2020	A*	`conf`	GeoDA: A Geometric Framework for Black-Box Adversarial Attacks. CVPR Ali Rahmati, Seyed-Mohsen Moosavi-Dezfooli, Pascal Frossard, Huaiyu Dai
2020	J	`jnl`	GeoDA: a geometric framework for black-box adversarial attacks. CoRR Ali Rahmati, Seyed-Mohsen Moosavi-Dezfooli, Pascal Frossard, Huaiyu Dai
2020	A*	`conf`	Hold me tight! Influence of discriminative features on deep network boundaries. NeurIPS Guillermo Ortiz-Jiménez, Apostolos Modas, Seyed-Mohsen Moosavi-Dezfooli, Pascal Frossard
2020	J	`jnl`	Hold me tight! Influence of discriminative features on deep network boundaries. CoRR Guillermo Ortiz-Jiménez, Apostolos Modas, Seyed-Mohsen Moosavi-Dezfooli, Pascal Frossard
2020	A*	`conf`	Neural Anisotropy Directions. NeurIPS Guillermo Ortiz-Jiménez, Apostolos Modas, Seyed-Mohsen Moosavi-Dezfooli, Pascal Frossard
2020	J	`jnl`	Neural Anisotropy Directions. CoRR Guillermo Ortiz-Jiménez, Apostolos Modas, Seyed-Mohsen Moosavi-Dezfooli, Pascal Frossard
2020	J	`jnl`	Optimism in the Face of Adversity: Understanding and Improving Deep Learning through Adversarial Robustness. CoRR Guillermo Ortiz-Jiménez, Apostolos Modas, Seyed-Mohsen Moosavi-Dezfooli, Pascal Frossard
2019	A*	`conf`	A Geometry-Inspired Decision-Based Attack. ICCV Yujia Liu, Seyed-Mohsen Moosavi-Dezfooli, Pascal Frossard
2019	J	`jnl`	A geometry-inspired decision-based attack. CoRR Yujia Liu, Seyed-Mohsen Moosavi-Dezfooli, Pascal Frossard
2019	—	`—`	Geometry of adversarial robustness of deep networks: methods and applications. Seyed-Mohsen Moosavi-Dezfooli
2019	A*	`conf`	Robustness via Curvature Regularization, and Vice Versa. CVPR Seyed-Mohsen Moosavi-Dezfooli, Alhussein Fawzi, Jonathan Uesato, Pascal Frossard
2019	A*	`conf`	SparseFool: A Few Pixels Make a Big Difference. CVPR Apostolos Modas, Seyed-Mohsen Moosavi-Dezfooli, Pascal Frossard
2019	Misc	`conf`	Universal Adversarial Attacks on Text Classifiers. ICASSP Melika Behjati, Seyed-Mohsen Moosavi-Dezfooli, Mahdieh Soleymani Baghshah, Pascal Frossard
2018	A*	`conf`	Adaptive Quantization for Deep Neural Network. AAAI Yiren Zhou, Seyed-Mohsen Moosavi-Dezfooli, Ngai-Man Cheung, Pascal Frossard
2018	J	`jnl`	Divide, Denoise, and Defend against Adversarial Attacks. CoRR Seyed-Mohsen Moosavi-Dezfooli, Ashish Shrivastava, Oncel Tuzel
2018	A*	`conf`	Empirical Study of the Topology and Geometry of Deep Networks. CVPR Alhussein Fawzi, Seyed-Mohsen Moosavi-Dezfooli, Pascal Frossard, Stefano Soatto
2018	A*	`conf`	Geometric Robustness of Deep Networks: Analysis and Improvement. CVPR Can Kanbak, Seyed-Mohsen Moosavi-Dezfooli, Pascal Frossard
2018	—	`conf`	Robustness of Classifiers to Universal Perturbations: A Geometric Perspective. ICLR (Poster) Seyed-Mohsen Moosavi-Dezfooli, Alhussein Fawzi, Omar Fawzi, Pascal Frossard, Stefano Soatto
2018	J	`jnl`	Robustness via curvature regularization, and vice versa. CoRR Seyed-Mohsen Moosavi-Dezfooli, Alhussein Fawzi, Jonathan Uesato, Pascal Frossard
2018	J	`jnl`	SparseFool: a few pixels make a big difference. CoRR Apostolos Modas, Seyed-Mohsen Moosavi-Dezfooli, Pascal Frossard
2017	J	`jnl`	Adaptive Quantization for Deep Neural Network. CoRR Yiren Zhou, Seyed-Mohsen Moosavi-Dezfooli, Ngai-Man Cheung, Pascal Frossard
2017	J	`jnl`	Analysis of universal adversarial perturbations. CoRR Seyed-Mohsen Moosavi-Dezfooli, Alhussein Fawzi, Omar Fawzi, Pascal Frossard, Stefano Soatto
2017	J	`jnl`	Classification regions of deep neural networks. CoRR Alhussein Fawzi, Seyed-Mohsen Moosavi-Dezfooli, Pascal Frossard, Stefano Soatto
2017	J	`jnl`	Geometric robustness of deep networks: analysis and improvement. CoRR Can Kanbak, Seyed-Mohsen Moosavi-Dezfooli, Pascal Frossard
2017	J	`jnl`	The Robustness of Deep Networks: A Geometrical Perspective. IEEE Signal Process. Mag. Alhussein Fawzi, Seyed-Mohsen Moosavi-Dezfooli, Pascal Frossard
2017	A*	`conf`	Universal Adversarial Perturbations. CVPR Seyed-Mohsen Moosavi-Dezfooli, Alhussein Fawzi, Omar Fawzi, Pascal Frossard
2016	A*	`conf`	DeepFool: A Simple and Accurate Method to Fool Deep Neural Networks. CVPR Seyed-Mohsen Moosavi-Dezfooli, Alhussein Fawzi, Pascal Frossard
2016	—	`conf`	Robustness of classifiers: from adversarial to random noise. NIPS Alhussein Fawzi, Seyed-Mohsen Moosavi-Dezfooli, Pascal Frossard
2016	J	`jnl`	Robustness of classifiers: from adversarial to random noise. CoRR Alhussein Fawzi, Seyed-Mohsen Moosavi-Dezfooli, Pascal Frossard
2016	B	`conf`	Simultaneous Acoustic Localization of Multiple Smartphones with Euclidean Distance Matrices. EWSN Seyed-Mohsen Moosavi-Dezfooli, Yvonne-Anne Pignolet, Dacfey Dzung
2016	J	`jnl`	Universal adversarial perturbations. CoRR Seyed-Mohsen Moosavi-Dezfooli, Alhussein Fawzi, Omar Fawzi, Pascal Frossard
2015	J	`jnl`	DeepFool: a simple and accurate method to fool deep neural networks. CoRR Seyed-Mohsen Moosavi-Dezfooli, Alhussein Fawzi, Pascal Frossard
2015	J	`jnl`	Simultaneous Acoustic Localization of Multiple Smartphones with Euclidean Distance Matrices. CoRR Seyed-Mohsen Moosavi-Dezfooli, Yvonne-Anne Pignolet, Dacfey Dzung