{"type": "article", "title": "MileGPO: Milestone Inference with Local Evidence for Graph-Based Policy Optimization of Long-Horizon LLM Agents", "publisher": "Web Pulse", "url": "https://wpnews.pro/news/milegpo-milestone-inference-with-local-evidence-for-graph-based-policy-of-long", "original_source": "https://www.machinebrief.com/news/milegpo-milestone-inference-with-local-evidence-for-graph-ba-cmm4", "published": "2026-08-21T04:00:00+00:00", "accessed": "2026-08-21", "id": "milegpo-milestone-inference-with-local-evidence-for-graph-based-policy-of-long"}