Building My First RAG System: Grounding LLMs in RealityOne of the first things you learn about large language models is their “knowledge cutoff.” Ask a model about an event that happened yesterday, and it will politely tell you it doesn’t have access to real-time information. For a project last semester, we were tasked with building a Q&A bot about recent developments in our field, and this limitation was a huge roadblock. That’s when our professor introduced us to Retrieval-Augmented Gen... Read article

How I Shrank My LLM: A Student’s Dive into Model QuantizationOne of the most exciting and frustrating moments in my Master’s program was when I finally got my hands on a powerful, pre-trained language model. The excitement came from its incredible capabilities; the frustration came when I realized it was too big to run on my university-provided GPU for any serious fine-tuning. This sent me down the rabbit hole of model compression, and my first major stop was quantization. What Exactly is Qua... Read article

A Grad Student’s Guide to Fine-Tuning LLMs: From Brute Force to FinesseWhen I was assigned my first big research project, the goal was to adapt a general-purpose large language model for a very specific task: analyzing sentiment in financial news. My first thought was, “Easy, I’ll just fine-tune it.” I quickly learned that “just fine-tuning” is a massive oversimplification. The journey taught me a ton about the different strategies we have at our disposal. The Default: Full-Parameter Fine-Tun... Read article

On Data Contamination in LLMs: A Grad Student’s PerspectiveIn my NLP seminar last semester, a recurring theme was the integrity of our evaluation benchmarks. We spent weeks discussing how to measure progress, but one topic that really stuck with me was data contamination—the subtle, almost accidental way we can end up “cheating” on our tests. It’s a problem that seems technical on the surface but cuts to the very core of our field’s credibility. The Core of the Problem: When Test Data Becomes... Read article