{"type":"link","version":"1.0","title":"Structured SFT on curated trajectories matches RL-trained agents on tool routing without reward shaping","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/agentic-behavior-distillation-sft-matches-rl-on-tool-routing","thumbnail_url":"https://ai-arch.pages.dev/og/agentic-behavior-distillation-sft-matches-rl-on-tool-routing.png","thumbnail_width":1200,"thumbnail_height":630}