{"slug": "neohorse-1-towards-recursive-self-improvement-via-agentic-post-training-with", "title": "NeoHorse-1: Towards Recursive Self-Improvement via Agentic Post-Training with Routing Harness", "summary": "Researchers introduced NeoHorse-1, a family of agent-native models designed for recursive self-improvement through agentic post-training with a routing harness. The system observes its own capabilities and converts that evidence into subsequent learning rounds, marking a step toward autonomous AI improvement.", "body_md": "Recursive self-improvement (RSI) requires a concrete mechanism through which an AI system observes its capabilities and converts that evidence into the next round of learning. We present NeoHorse-1, a family of agent-native models developed to explore this path through agentic post-training. Our sys", "url": "https://wpnews.pro/news/neohorse-1-towards-recursive-self-improvement-via-agentic-post-training-with", "canonical_source": "https://aiflash.com/news/116073/", "published_at": "2026-09-09 03:00:36+00:00", "updated_at": "2026-09-09 03:20:34.714682+00:00", "lang": "en", "topics": ["artificial-intelligence", "ai-research", "ai-agents", "machine-learning"], "entities": ["NeoHorse-1"], "alternates": {"html": "https://wpnews.pro/news/neohorse-1-towards-recursive-self-improvement-via-agentic-post-training-with", "markdown": "https://wpnews.pro/news/neohorse-1-towards-recursive-self-improvement-via-agentic-post-training-with.md", "text": "https://wpnews.pro/news/neohorse-1-towards-recursive-self-improvement-via-agentic-post-training-with.txt", "jsonld": "https://wpnews.pro/news/neohorse-1-towards-recursive-self-improvement-via-agentic-post-training-with.jsonld"}}