Stories by fofoz

Global Optimization of Black-Box Functions with Unknown Lipschitz Constants

Understanding Why Adam Outperforms SGD: Gradient Heterogeneity in Transformers

Sundial: A Family of Highly Capable Time Series Foundation Models

DeepRAG: Thinking to retrieval step by step for large language models

Language Models Use Trigonometry to Do Addition

OmniHuman-1: Human Animation Models

Efficient Reasoning with Hidden Thinking

WeatherNext

DeepSeek-R1 models now available on AWS

SFT Memorizes,RL Generalizes: Comparative Study of Foundation Model PostTraining

Engineered heart muscle allografts for heart repair in primates and humans

2024 YR4

Janus Pro 1B running 100% locally in-browser on WebGPU

YuE: Open Full-Song Generation Foundation Model

Sky-T1: Train your own O1 preview model within $450

Rust 1.84

The Year in Math

The Year in Computer Science

Video Representation Learning with Joint-Embedding Predictive Architectures

1 2 3 4 5 6