Agentic Knowledgeable Self-awareness

摘要

Large Language Models (LLMs) have achieved considerable performance across various agentic planning tasks. However, traditional approaches adopt a “flood irrigation” methodology that indiscriminately injects gold trajectories, external feedback, and domain knowledge into agent models. This practice overlooks the fundamental human cognitive principle of self-awareness - the ability to dynamically assess situational demands and strategically employ resources during decision-making. We propose \textbf{Agentic Knowledgeable Self-awareness} to address this gap, a novel paradigm enabling LLM-based agents to autonomously regulate knowledge utilization. Specifically, we propose \texttt{KnowSelf}, a data-centric approach that applies agents with \texttt{know}ledgeable \texttt{self}-awareness like humans. Concretely, we devise a heuristic situation judgement criterion to mark special tokens on the agent’s self-explored trajectories for collecting training data. Through a two-stage training process, the agent model can switch between different situations by generating specific special tokens, achieving optimal planning effects with minimal costs. Our experiments demonstrate that can outperform various strong baselines on different tasks and models with minimal use of external knowledge.