用验证器生成的正确决策训练智能体，让规划系统更便宜更快。
Training the Orchestrator: A Supervised Approach to End-to-End PDDL Planning with LLM Agents
- 用验证通过的规划轨迹监督训练调度器，替代每步调用大模型。
- 在多个基准上成功率媲美大模型，成本降低45倍以上。
- 适合需要高效、低成本智能规划的工业应用或研究场景。
将自然语言规划意图转化为可验证计划是长期挑战：人类用语言表达目标，而经典规划器需要形式化的PDDL规范。近期的智能体框架通过在验证器检查的迭代优化循环中协调一组专用修复智能体来弥合这一差距，但核心调度器本身仍是一个提示驱动的前沿大模型，在每一步优化中都要调用前沿大模型API。我们提出HALO（混合代理-学习调度器），通过外部验证器认证为有效计划的优化轨迹，对11个PDDL领域中的调度器进行训练。HALO结合一个小规模QLoRA微调策略与三条硬编码规则以处理可直接判定的选择，并作用于扩展后的21个智能体动作空间。不同于每步都提示前沿大模型或仅从任务结束时稀疏奖励中学习的方法，我们的关键观察是：验证器已提供强指导——每个被接受的轨迹都是可证明正确的（状态, 智能体）决策序列，可直接作为监督信号。在PlanBench、Natural Plan及经典规划基准上，HALO在成功率上达到或超过使用GPT-5-mini提示的基线，与更强的Gemini-3-Flash提示基线相差不到3个百分点，调度成本降低超过一个数量级（从每任务$0.18降至$0.004，约45倍便宜；比Gemini-3-Flash便宜约15倍），每轮总大模型调用减少40%至50%。
原文摘要 · Abstract (English)
Translating natural-language planning intent into verified plans is a longstanding challenge: people communicate goals in language, while classical planners require formal PDDL specifications. Recent agentic frameworks bridge this gap by orchestrating a pool of specialized repair agents inside a verifier-checked refinement loop, but the orchestrator at the centre is itself a prompted frontier LLM, paying a frontier-LLM API call at every refinement step. We present HALO (Hybrid Agent-Learned Orchestrator), which trains the orchestrator from refinement trajectories that an external verifier has certified as ending in valid plans, across 11 PDDL domains. HALO pairs a small QLoRA-tuned policy with three hardcoded rules for trivially decidable selections, and operates over an expanded 21-agent action space. Unlike approaches that prompt a frontier LLM at every step or learn an orchestrator from sparse end-of-episode rewards, our key observation is that the verifier already provides strong guidance: every accepted trajectory is a sequence of demonstrably correct (state, agent) decisions, directly usable as supervision. Across PlanBench, Natural Plan, and classical planning benchmarks, HALO matches or exceeds the GPT-5-mini prompted baseline on success rate, sits within three percentage points of the stronger Gemini-3-Flash prompted baseline, reduces orchestration cost by more than an order of magnitude (\$0.18 to \$0.004 per task against GPT-5-mini, roughly 45$\times$ cheaper; roughly 15$\times$ cheaper than Gemini-3-Flash), and cuts total LLM calls per episode by 40 to 50 percent.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。