Research

Academic Work

Publications and reading notes on AI, computer vision, and deep learning. Research interests span generative models, medical imaging, and efficient inference.

Publications

IEEE Access · 2024Journal

A Comparative Study of State-of-the-Art Deep Learning Models for Semantic Segmentation of Pores in SEM Images

B. Pokharel, B. Yadav, et al. · vol. 12, pp. 50217–50243

A systematic comparison of modern deep learning architectures for semantic segmentation of pore structures in scanning electron microscope (SEM) images, with applications in materials science and quality control.

Computer VisionSemantic SegmentationSEMDeep Learning

Reading Notes

Key takeaways from papers I've read and found influential.

Attention Is All You Need

Vaswani et al. · 2017

Foundational transformer paper. The multi-head attention mechanism and positional encoding are still central to modern LLMs.

TransformersNLPAttention

An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale

Dosovitskiy et al. · 2020

ViT showed that pure transformers can match CNNs on image tasks when pre-trained on large datasets. Key insight: patch embeddings bridge vision and NLP.

ViTComputer VisionTransformers

LLaMA: Open and Efficient Foundation Language Models

Touvron et al. · 2023

Demonstrated that smaller models trained on more data can outperform larger ones. Sparked the open-source LLM ecosystem.

LLMOpen SourceEfficiency