Papershelf
Some Reserach papers that I found amusing

Why Language Models Hallucinate

OpenAI, 2025

LLMs are known to produce overconfident, plausible falsehoods, which diminish their utility and trustworthiness. This error mode is known as “hallucination

LLM Algebra

The Illusion of Thinking

Apple, 2025

Understanding the Strengths and Limitations of Reasoning Models

Transformers LLM Limitations

Dismissing Python Garbage Collection at Instagram

Instagram, 2017

By disabling Garbage Collection, we can reduce the memory footprint

Python Memory

Twitter’s in-house photo storage system

Twitter, 2017

Blobstore, low-cost and scalable storage system built to store photos and other binary large objects, also known as blobs

Data structures Backend