{"type":"link","version":"1.0","title":"Critic-free RL fine-tuning that uses group reward statistics as the baseline","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/feat-critic-free-group-baseline-rl-trainer","thumbnail_url":"https://ai-arch.pages.dev/og/feat-critic-free-group-baseline-rl-trainer.png","thumbnail_width":1200,"thumbnail_height":630}