How to Use LLMs for Powerful Automatic Evaluations

discuss how you can perform automatic evaluations using LLM as a judge. LLMs are widely used today for a variety ...
Read more
Coconut: A Framework for Latent Reasoning in LLMs

Paper link: https://arxiv.org/abs/2412.06769 Released: 9th of December 2024 Figure 1. The two reasoning modes of Coconut. In Language Mode (left), ...
Read more
How to Write Insightful Technical Articles

, I discuss how you can write technical articles. I have been writing such articles for around 2.5 years, and ...
Read more
How to Design Machine Learning Experiments — the Right Way

Never miss a new edition of The Variable, our weekly newsletter featuring a top-notch selection of editors’ picks, deep dives, ...
Read more
From Data Scientist IC to Manager: One Year In

year, I wrote about my first 3-month manager experience. I shared some of the immediate changes I noticed, including more ...
Read more
When Models Stop Listening: How Feature Collapse Quietly Erodes Machine Learning Systems

A was implemented, studied, and proved. It was right in its predictions, and its metrics were consistent. The logs were ...
Read more
How Computers “See” Molecules | Towards Data Science

a computer, Edvard Munch’s The Scream is nothing more than a grid of pixel values. It has no sense of why swirling ...
Read more
Mastering NLP with spaCy – Part 2

in a sentence provide a lot of information, such as what they mean in the real world, how they connect ...
Read more
The Misconception of Retraining: Why Model Refresh Isn’t Always the Fix

phrase “just retrain the model” is deceptively simple. It has become a go-to solution in machine learning operations whenever the ...
Read more