Explore >> Select a destination
|
You are here 
|
magazine.sebastianraschka.com | ||
| | | | |
transformer-circuits.pub
|
|
| | | | | ||
| | | | |
haifengl.wordpress.com
|
|
| | | | | Generative artificial intelligence (GenAI), especially ChatGPT, captures everyone's attention. The transformerbased large language models (LLMs), trained on a vast quantity of unlabeled data at scale, demonstrate the ability to generalize to many different tasks. To understand why LLMs are so powerful, we will deep dive into how they work in this post. LLM Evolutionary Tree... | |
| | | | |
www.shaped.ai
|
|
| | | | | This article explores how cross-encoders, long praised for their performance in neural ranking, may in fact be reimplementing classic information retrieval logic, specifically, a semantic variant of BM25. Through mechanistic interpretability techniques, the authors uncover circuits within MiniLM that correspond to term frequency, IDF, length normalization, and final relevance scoring. The findings bridge modern transformer-based relevance modeling with foundational IR principles, offering both theoretical insight and a roadmap for building more transparent and interpretable neural retrieval systems. | |
| | | | |
jaketae.github.io
|
|
| | | Recently, a friend recommended me a book, Deep Learning with Python by Francois Chollet. As an eager learner just starting to fiddle with the Keras API, I decided it was a good starting point. I have just finished the first section of Part 2 on Convolutional Neural Networks and image processing. My impression so far is that the book is more focused on code than math. The apparent advantage of this approach is that it shows readers how to build neural networks very transparently. It's also a good introduction to many neural network models, such as CNNs or LSTMs. On the flip side, it might leave some readers wondering why these models work, concretely and mathematically. This point notwithstanding, I've been enjoying the book very much so far, and this post is... | ||
