analiza aiqbench
This dissertation investigates critical security vulnerabilities in NLP and Vision-Language models, specifically focusing on backdoor attacks that threaten the integrity of Trustworthy AI systems. By analyzing detection mechanisms and designing countermeasures for multimodal representations, the work directly addresses the safety and robustness challenges inherent in deploying advanced deep learning architectures.
Backdoor Learning in Language Models and Vision-Language Models
arXiv:2608.18095v1 Announce Type: cross Abstract: Recent advances in deep learning have significantly enhanced the capabilities of Natural Language Processing (NLP) and Vision-Language Models (VLMs). However, these advancements come with increased vulnerabilities, notably through backdoor attacks that pose severe security threats. This thesis addresses two critical dimensions of Trustworthy AI and Efficient Multimodal Representation Learning: (1) security through analyzing, detecting, and designing backdoor attacks in NLP and VLMs, and (2) efficiency through advanced multimodal representation methods tailored for clinical and medical imaging applications.
Abstract:Recent advances in deep learning have significantly enhanced the capabilities of Natural Language Processing (NLP) and Vision-Language Models (VLMs). However, these advancements come with increased vulnerabilities, notably through backdoor attacks that pose severe security threats. This thesis addresses two critical dimensions of Trustworthy AI and Efficient Multimodal Representation Learning: (1) security through analyzing, detecting, and designing backdoor attacks in NLP and VLMs, and (2) efficiency through advanced multimodal representation methods tailored for clinical and medical imaging applications.
Ph.D. dissertation
Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
arXiv:2608.18095 [cs.CL]
(or arXiv:2608.18095v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2608.18095 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 175 cuvinte
Subiecte conexe
Architectural Backdoors in Vision-Language Model Supply Chains via Representation Steering
arXiv · 29 iul. 2026
BYORn: Bootstrap Your Own Responses to Defend Large Vision-Language Models Against Backdoor Attacks
arXiv · 3 iun. 2026
StepTrigger: Contact-State-Triggered Backdoor Attacks on VLM-Powered Legged Robots
arXiv · 24 sept. 2026
State Backdoor: Towards Stealthy Real-world Poisoning Attack on Vision-Language-Action Model in State Space
arXiv · 9 iun. 2026