7 papers
A Self-Explainable Deep Architecture for Security Applications
Ananth Shreekumar, Jyun-Jhu Syu, Muslum Ozgur Ozmen +2
Deep learning models have become integral to security applications due to their ability to model complex relationships in data and detect sophisticated threats. However, their comp…
The Yes-Man Syndrome: Benchmarking Abstention in Embodied Robotic Agents
Doguhan Yeke, Elif Su Temirel, Ananth Shreekumar +3
Vision-language models (VLMs) are used as high-level planners for embodied agents, translating natural language instructions and visual observations into action plans. While prior…
RoboJailBench: Benchmarking Adversarial Attacks and Defenses in Embodied Robotic Agents
Doguhuan Yeke, Yanming Zhou, Leo Y. Lin +3
Recent advances in Vision-Language Models (VLMs) facilitate a new class of embodied AI systems, where these models are integrated into physical platforms, e.g. robots and autonomou…
Stable GFlowNets with TV Monitoring and Probabilistic Guarantees
Zengxiang Lei, Ananth Shreekumar, Jonathan Rosenthal +6
Generative Flow Networks (GFlowNets) sample diverse structured objects in proportion to reward and have been applied to molecular discovery and biological-sequence design, where fi…
Formalizing the Safety, Security, and Functional Properties of Agentic AI Systems
Edoardo Allegrini, Ananth Shreekumar, Z. Berkay Celik
Agentic AI systems, which leverage multiple autonomous agents and large language models (LLMs), are increasingly used to address complex, multi-step tasks. The safety, security, an…
STARS: Synchronous Token Alignment for Robust Supervision in Large Language Models
Mohammad Atif Quamar, Mohammad Areeb, Mikhail Kuznetsov +2
Aligning large language models (LLMs) with human values is crucial for safe deployment. Inference-time techniques offer granular control over generation; however, they rely on mode…