Active Epistemic Control for Query-Efficient Verified Planning

TL;DR

This paper introduces Active Epistemic Control (AEC), a planning layer that manages belief and commitment in environments with partial observability, improving efficiency and reducing replanning in complex tasks.

Contribution

The paper proposes AEC, a novel epistemic-categorical planning approach that separates belief management from commitment, integrating environment querying and simulation for better planning under uncertainty.

Findings

AEC achieves competitive success on ALFWorld and ScienceWorld.

AEC requires fewer replanning rounds than strong LLM-agent baselines.

The approach effectively balances querying and simulation to handle uncertainty.

Abstract

Planning in interactive environments is challenging under partial observability: task-critical preconditions (e.g., object locations or container states) may be unknown at decision time, yet grounding them through interaction is costly. Learned world models can cheaply predict missing facts, but prediction errors can silently induce infeasible commitments. We present \textbf{Active Epistemic Control (AEC)}, an epistemic-categorical planning layer that integrates model-based belief management with categorical feasibility checks. AEC maintains a strict separation between a \emph{grounded fact store} used for commitment and a \emph{belief store} used only for pruning candidate plans. At each step, it either queries the environment to ground an unresolved predicate when uncertainty is high or predictions are ambiguous, or simulates the predicate to filter hypotheses when confidence is sufficient. Final commitment is gated by grounded precondition coverage and an SQ-BCP pullback-style compatibility check, so simulated beliefs affect efficiency but cannot directly certify feasibility. Experiments on ALFWorld and ScienceWorld show that AEC achieves competitive success with fewer replanning rounds than strong LLM-agent baselines.