SCOPE: Evolving Symbolic World for Planning in Open-Ended Environments

SCOPE is a new self-adaptive symbolic planning framework designed for open-ended environments, integrating a Symbolic Execution Simulator (SESim) and a Self-Adaptive Symbolic Memory (SASMem) to enhance planning through iterative refinement of symbolic representations. The framework addresses the challenges of incomplete symbolic worlds by enabling real-time validation and adaptation of action plans, significantly improving planning success rates and adaptability across various tasks. This development is crucial for practitioners as it enhances the robustness of planning systems in dynamic and uncertain environments, facilitating more effective long-horizon task execution.