AegisLLM: Scaling Agentic Systems for Self-Reflective Defense in LLM Security.
CoRR, April, 2025
Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilities.
,
,
,
,
,
,
,
,
,
,
,
,
,
,
CoRR, February, 2025
Learning to Attack, Protect, and Enhance Deep Networks
PhD thesis, 2024
Single Layer Single Gradient Unlearning.
CoRR, 2024
Transformation-Dependent Adversarial Attacks.
CoRR, 2024
Cross-Modal Safety Alignment: Is textual unlearning all you need?
CoRR, 2024
Can Textual Unlearning Solve Cross-Modality Safety Alignment?
Proceedings of the Findings of the Association for Computational Linguistics: EMNLP 2024, 2024
Disguise without Disruption: Utility-Preserving Face De-identification.
Proceedings of the Thirty-Eighth AAAI Conference on Artificial Intelligence, 2024
Ensemble-based Blackbox Attacks on Dense Prediction.
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023
Learning to Sense for Coded Diffraction Imaging.
Sensors, 2022
Blackbox Attacks via Surrogate Ensemble Search.
CoRR, 2022
Blackbox Attacks via Surrogate Ensemble Search.
Proceedings of the Advances in Neural Information Processing Systems 35: Annual Conference on Neural Information Processing Systems 2022, 2022
Zero-Query Transfer Attacks on Context-Aware Object Detectors.
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022
Context-Aware Transfer Attacks for Object Detection.
Proceedings of the Thirty-Sixth AAAI Conference on Artificial Intelligence, 2022
Data-Driven Illumination Patterns For Coded Diffraction Imaging.
Proceedings of the 2021 IEEE International Conference on Image Processing, 2021
Exploiting Multi-Object Relationships for Detecting Adversarial Attacks in Complex Scenes.
Proceedings of the 2021 IEEE/CVF International Conference on Computer Vision, 2021
Solving Phase Retrieval with a Learned Reference.
Proceedings of the Computer Vision - ECCV 2020, 2020