←Backuclaml/SPIN0Copy as MarkdownView on GitHub↗1,245 stars·105 forks·Python·Apache-2.0·0 viewsuclaml.github.io/SPIN↗SPINFeaturesReinforcement Learning - Self-play fine-tuning to enhance language model performance.Self-Improvement Methods - Uses self-play to refine language models without external human data.