machine learning - Abrarqasim Blogs

LLM Evaluation Metrics: What I Actually Trust After Judges Failed Me

I ran an LLM judge for six weeks before I noticed it kept rewarding longer answers. Here is the evaluation metrics stack I actually trust…

Fine-tune an LLM and a small open model can beat GPT-5 on a narrow task, cheaper. A look at when QLoRA wins, when it loses,…