5 papers
Are LLM Uncertainty and Correctness Encoded by the Same Features? A Functional Dissociation via Sparse Autoencoders
Het Patel, Tiejin Chen, Hua Wei +2
Large language models can be uncertain yet correct, or confident yet wrong, raising the question of whether their output-level uncertainty and their actual correctness are driven b…
Extracting and Analyzing Rail Crossing Behavior Signatures from Videos using Tensor Methods
Dawon Ahn, Het Patel, Aemal Khattak +2
Railway crossings present complex safety challenges where driver behavior varies by location, time, and conditions. Traditional approaches analyze crossings individually, limiting…
AD-SAM: Fine-Tuning the Segment Anything Vision Foundation Model for Autonomous Driving Perception
Mario Camarena, Het Patel, Fatemeh Nazari +3
This paper presents the Autonomous Driving Segment Anything Model (AD-SAM), a fine-tuned vision foundation model for semantic segmentation in autonomous driving (AD). AD-SAM extend…
Robust Vision-Language Models via Tensor Decomposition: A Defense Against Adversarial Attacks
Het Patel, Muzammil Allie, Qian Zhang +2
Vision language models (VLMs) excel in multimodal understanding but are prone to adversarial attacks. Existing defenses often demand costly retraining or significant architecture c…
TRAWL: Tensor Reduced and Approximated Weights for Large Language Models
Yiran Luo, Het Patel, Yu Fu +4
Recent research has shown that pruning large-scale language models for inference is an effective approach to improving model efficiency, significantly reducing model weights with m…