Inherent Vulnerabilities of Vision Transformers for Medical Image Classification
Explored vulnerabilities in vision transformers for medical image classification, demonstrating that minor modifications can significantly alter representations, and developed an efficient detection method to enhance model reliability.
Mar 2, 2024
Deciphering Poorly Formatted Text using Private LLM
Process unstructured text data from document files efficiently, prioritizing data privacy with the implementation of locally hosted private large language models. Employed a PEFT model (LoRA) to optimize finetuning for the summarizer and classifier, yielding improved performance metrics with the help of prompt engineering and LLM reasoning.
Apr 25, 2023