view article Article Tokenization in Transformers v5: Simpler, Clearer, and More Modular +4 Dec 18, 2025 • 116
Running on CPU Upgrade Featured 2.92k The Smol Training Playbook 📚 2.92k The secrets to building world-class LLMs
Running 3.66k The Ultra-Scale Playbook 🌌 3.66k The ultimate guide to training LLM on large GPU Clusters
VR-Thinker: Boosting Video Reward Models through Thinking-with-Image Reasoning Paper • 2510.10518 • Published Oct 12, 2025 • 19
TokDrift: When LLM Speaks in Subwords but Code Speaks in Grammar Paper • 2510.14972 • Published Oct 16, 2025 • 35
TokDrift: When LLM Speaks in Subwords but Code Speaks in Grammar Paper • 2510.14972 • Published Oct 16, 2025 • 35