I am a Senior majoring in Computer Science at Auburn University who's passionate about AI/ML and improving our understanding of the world through science. I have experience with Python, PyTorch, SQL, JavaScript, C#, HTML/CSS and more.
View ResumeThis paper examines the limitations of current vision-based language models, such as GPT-4o and Sonnet 3.5, in performing low-level vision tasks. Despite their high scores on numerous multimodal benchmarks, these models often fail on very basic cases. This raises the question: are we evaluating these models accurately?
This paper explores how allowing LLMs to highlight their chain of thought to visually link information in the question and answer can increase LLM accuracy while also improving user experience. (arxiv preprint coming soon)
Classified if tweets were referencing real world disasters or not using a fine tuned BERT model. Achieved 83% accuracy and top 10% performance amongst thousands of competitors in a Kaggle competition.
Developed a YOLOv5 model for facial recognition using a publicly available dataset of 1,000+ images resulting in a model that is able to quickly recognize faces in real-time videos.
Created a neural network from scratch using Python and Numpy without the use of modern libaries like Pytorch. The model was trained on the MNIST dataset and achieved 92% accuracy using only a shallow network.
Have a question or want to work together? Feel free to send me an email:
loganbolton101@gmail.com