Introductory article on how Columnar Databases work.

Why do column oriented databases prefer Delete-Insert over Update?

In this article, we will be particularly focussing on **Why is Heap allocation slow?**. We will look at operations that are needed while allocating heap memory and we will try to understand them properly.

Why Is Heap Allocation Slower than Stack Allocation?

A superficial intro to the actual hardware components that we have discussed perviously.

What happens when CPU wants to modify the data?

Hardware Case Study

All the Loop Optimizations techniques that arise from Cache Hierarchy and Miss rate reduction.

Loop Optimizations

The compiler level optimizations that are related to and affected by the `volatile` keyword. How they work and their implications.

Intro to Common Subexpression Elimination

The Volatile Keyword in Cpp

Let's dive deeper into how shifting and multiplication operations work on lower-level.

2's complement representation of binary numbers -

Multiplication - `imul` and Booth's algorithm -

Why is shifting faster than multiplication?

This post simulates guaranteed & in-order delivery. Which are a feature of TCP and are lacking in UDP.

What is Data Sequencing (In-order transmission)

Simulating Guaranteed & In-order delivery

More often than not, I have heard people mentioning that while calling a function with some arguments, `Pass By Reference` is a lot faster than `Pass By Value`. Lets check if this is a solid claim.

Is Pass By Reference really that fast?

So this Pass By Value vs Pass By Reference is really a big confusing part of day to day programming. And when you get into some Functional Programming-oriented environments like ReactJS, this thing just tops the list of annoying things.

Pass By Value vs Reference! What's the default really?

Abstract - We expose a surprising failure of generalization in auto-regressive large language models (LLMs). If a model is trained on a sentence of the form “A is B”, it will not automatically generalize to the reverse direction “B is A”. This is the Reversal Curse. For instance, if a model is trained on “Valentina Tereshkova was the first woman to travel to space”, it will not automatically be able to answer the question, “Who was the first woman to travel to space?”. Moreover, the likelihood of the correct answer (“Valentina Tershkova”) will not be higher than for a random name.

The Reversal Curse- LLMs Trained On “A IS B” Fail To Learn "B IS A"

Excel FlashFill original paper. Textbook example for learning Program Synthesis.

Automating String Processing in Spreadsheets Using Input-Output Examples

One of the first papers to introduce embeddings as a concept, and neural networks as a general solution to language modelling.

/ papershelf

/ blogs