KEEP: A KV-Cache-Centric Memory Management System for Efficient Embodied Planning

Abstract

Memory-augmented Large Language Models (LLMs) have demonstrated remarkable capability for complex and long-horizon embodied planning. By keeping track of past experiences and environmental states, memory enables LLMs to maintain a global view, thereby avoiding repetitive exploration. However, existing approaches often store memory as raw text, leading to excessively long prompts and high prefill latency. While it is possible to store and reuse KV caches, the efficiency benefits are greatly undermined due to frequent KV cache updates. In this paper, we propose KEEP, a KV-cache-centric memory management system for efficient embodied planning. KEEP features three key innovations: (1) a Static-Dynamic Memory Construction algorithm that reduces KV cache recomputation by mixed-granularity memory group; (2) a Multi-hop Memory Re-computation algorithm that dynamically identifies important cross-attention among different memory groups and reconstructs memory interactions iteratively; (3) a Layer-balanced Memory Loading that eliminates unbalanced KV cache loading and cross-attention computation across different layers. Extensive experimental results have demonstrated that KEEP achieves 2.68x speedup with negligible accuracy loss compared with text-based memory methods on the ALFRED dataset. Compared with the KV re-computation method CacheBlend (EuroSys 2025), KEEP shows 4.13% success rate improvement and 1.90x time-to-first-token (TTFT) reduction. Our code is available at https://github.com/PKU-SEC-Lab/KEEP_Embodied_Memory.

Recommended citation: Zebin Yang, Tong Xie, Baotong Lu, Shaoshan Liu, Bo Yu, and Meng Li. (2026). "KEEP: A KV-Cache-Centric Memory Management System for Efficient Embodied Planning." DAC.