{"type":"link","version":"1.0","title":"Sample from the model, rank by unsupervised signal, fine-tune on winners—a closed loop that avoids external expert models","author_name":"AI Archs","author_url":"https://ai-arch.pages.dev","provider_name":"AI Archs","provider_url":"https://ai-arch.pages.dev","url":"https://ai-arch.pages.dev/n/sampling-and-unsupervised-ranking-enables-self-improvement","thumbnail_url":"https://ai-arch.pages.dev/og/sampling-and-unsupervised-ranking-enables-self-improvement.png","thumbnail_width":1200,"thumbnail_height":630}