{
  "id": 17039,
  "url": "https://arxiv.org/abs/2608.05466",
  "title": "Recursive Synthesis for Long-Horizon Terminal Tasks",
  "summary": "High-quality long-horizon training data for terminal agents is expensive to produce, often costing hundreds to thousands of dollars per task, because each task must keep the instruction, environment, reference solution, and verifier mutually consistent. Human authoring does not scale, and direct generation with large language models (LLMs) often breaks these dependencies. We present Recursive Synthetic Terminal Tasks (RST), a recursive verified synthesis framework for constructing long-horizon t",
  "authors": "Zhongzhi Li, Yucheng Shi, Zongxia Li, Ruhan Wang, Anhao Li, Zixun Huang",
  "category": "research",
  "topics": "agents-autonomy,environment",
  "orgs": null,
  "regions": null,
  "published_at": "2026-08-04T20:00:00.000Z",
  "fetched_at": "2026-08-07T05:10:58.501Z",
  "source_slug": "hf-daily",
  "source_name": "HuggingFace Daily Papers",
  "source_homepage": "https://huggingface.co/papers",
  "ethics_ai_record_url": "https://ethics.ai/record/17039",
  "original_url": "https://arxiv.org/abs/2608.05466",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}