🤖 AI Summary
To address the limited planning and decision-making capabilities of large language models (LLMs) in complex, adversarial, long-horizon environments, this paper proposes PLAP—a parametric skill-driven hierarchical planning framework. PLAP comprises an LLM-based skill planner, an environment-specific parametric skill library, and a skill executor, enabling end-to-end mapping from natural-language instructions to reliable action sequences while reducing reliance on handcrafted rules and action reliability heuristics. Evaluated on MicroRTS, GPT-4o–driven PLAP achieves zero-shot performance surpassing 80% of baselines; Qwen2-72B with few-shot prompting outperforms the top scripted agent CoacAI. Furthermore, the authors introduce the first LLM benchmark leaderboard dedicated to long-horizon skill-based planning, establishing a standardized evaluation framework for this emerging research direction.
📝 Abstract
Recent advancements in Large Language Models(LLMs) have led to the development of LLM-based AI agents. A key challenge is the creation of agents that can effectively ground themselves in complex, adversarial long-horizon environments. Existing methods mainly focus on (1) using LLMs as policies to interact with the environment through generating low-level feasible actions, and (2) utilizing LLMs to generate high-level tasks or language guides to stimulate action generation. However, the former struggles to generate reliable actions, while the latter relies heavily on expert experience to translate high-level tasks into specific action sequences. To address these challenges, we introduce the Plan with Language, Act with Parameter (PLAP) planning framework that facilitates the grounding of LLM-based agents in long-horizon environments. The PLAP method comprises three key components: (1) a skill library containing environment-specific parameterized skills, (2) a skill planner powered by LLMs, and (3) a skill executor converting the parameterized skills into executable action sequences. We implement PLAP in MicroRTS, a long-horizon real-time strategy game that provides an unfamiliar and challenging environment for LLMs. The experimental results demonstrate the effectiveness of PLAP. In particular, GPT-4o-driven PLAP in a zero-shot setting outperforms 80% of baseline agents, and Qwen2-72B-driven PLAP, with carefully crafted few-shot examples, surpasses the top-tier scripted agent, CoacAI. Additionally, we design comprehensive evaluation metrics and test 6 closed-source and 2 open-source LLMs within the PLAP framework, ultimately releasing an LLM leaderboard ranking long-horizon skill planning ability. Our code is available at https://github.com/AI-Research-TeamX/PLAP.