[NeurIPS 2025 Spotlight] TPA: Tensor ProducT ATTenTion Transformer (https://arxiv.org/abs/2501.06425)

unofficial implementation of MARS-AdamW in PyTorch

The official implementation of MARS: Unleashing the Power of Variance Reduction for Training Large Models

An automatic evaluator for instruction-following language models. Human-validated, high-quality, cheap, and fast.

The official implementation of Self-Play Preference Optimization (SPPO)

A complete computer science study plan to become a software engineer.