Paper-Replications: From-Scratch PyTorch Replications of Classic and SOTA AI/ML Papers
Paper-Replications is a collection of from-scratch PyTorch implementations replicating classic and state-of-the-art AI/ML papers: transformer variants, attention mechanisms, GANs, diffusion models, CLIP-style vision-language models, LoRA and MoE fine-tuning, and more. The Paper-Replications source code and notebooks ar...