Harel Lidar

AI Researcher & Machine Learning Engineer

I explore the frontiers of artificial intelligence through research and implementation. My work focuses on neural architectures, 3D vision, and transformer models. I write about papers I read and ideas I explore.

GitHub LinkedIn Email

Research Interests

Neural Architectures — Transformer variants, mixture-of-experts, efficient attention
3D Vision — Gaussian splatting, neural radiance fields, scene reconstruction
Language Models — Scaling laws, alignment (DPO, PPO), mechanistic interpretability
Efficient ML — Model compression, pruning, knowledge distillation

Recent Posts

View all posts →

Selected Writing

3D Gaussian Splatting for Real-Time Radiance Field Rendering 3D Vision
A Primer on the Inner Workings of Transformer-Based Language Models Transformers
Is DPO Superior to PPO for LLM Alignment? Alignment
The Mathematics and Philosophy Behind MSE Foundations

View all selected writing →