ReCAP: Recursive Context-Aware Reasoning and Planning for Large Language Model Agents

TL;DR

ReCAP is a hierarchical framework that enhances large language models' ability to perform long-horizon reasoning and planning by maintaining coherent context, reducing redundant prompts, and improving success rates on complex benchmarks.

Contribution

ReCAP introduces a novel recursive, context-aware reasoning and planning framework that addresses limitations of existing methods in long-horizon tasks for LLMs.

Findings

Significantly improves subgoal alignment and success rates on reasoning benchmarks.

Achieves 32% gain on synchronous Robotouille and 29% on asynchronous Robotouille.

Reduces context drift and runtime overhead in hierarchical prompting.

Abstract

Long-horizon tasks requiring multi-step reasoning and dynamic re-planning remain challenging for large language models (LLMs). Sequential prompting methods are prone to context drift, loss of goal information, and recurrent failure cycles, while hierarchical prompting methods often weaken cross-level continuity or incur substantial runtime overhead. We introduce ReCAP (Recursive Context-Aware Reasoning and Planning), a hierarchical framework with shared context for reasoning and planning in LLMs. ReCAP combines three key mechanisms: (i) plan-ahead decomposition, in which the model generates a full subtask list, executes the first item, and refines the remainder; (ii) structured re-injection of parent plans, maintaining consistent multi-level context during recursive return; and (iii) memory-efficient execution, bounding the active prompt so costs scale linearly with task depth. Together these mechanisms align high-level goals with low-level actions, reduce redundant prompting, and preserve coherent context updates across recursion. Experiments demonstrate that ReCAP substantially improves subgoal alignment and success rates on various long-horizon reasoning benchmarks, achieving a 32% gain on synchronous Robotouille and a 29% improvement on asynchronous Robotouille under the strict pass@1 protocol.