Back to AI News
arxiv

HeMeR: Heterogeneous Memory Reconciliation for Embodied Agents via Structured KV Reuse

HeMeR is an inference framework for embodied agents that reconciles key-value cache states across declarative, procedural, and working memory as those contents change during interaction. Instead of reprocessing each updated memory context as new input, it preserves KV states for unchanged memory and recomputes only the parts affected by updates. The authors evaluate it on ALFWorld, Habitat, and real-world robotic deployments. In the real-world setting, they report 83.20% task success and a reduction in per-step inference time from 2.94 to 1.87 seconds compared with MemRL.

AI Brief

It offers a concrete efficiency technique for LLM-based embodied agents, with reported latency and task-success results that are useful to researchers and builders working on memory-heavy agent inference.

arXiv:2611.16487v1 Announce Type: new Abstract: Embodied agents make decisions by combining information from heterogeneous memory modules, including declarative memory for explicit knowledge, procedural memory for learned routines, and working memory for current observations. During interaction, the memory contents required for decision making change with the agent's state, goal, and task progress, while many contents remain shared across consecutive decisions. However, existing memory modules are typically accessed through separate interfaces, so each updated memory context is often processed as a new input to the foundation model. This prevents selective key-value (KV) cache reuse for unchanged memory contents and forces full KV cache recomputation even when much of the memory context remains valid. We introduce HeMeR, an inference framework that performs KV reconciliation between dynamically updated memory contexts and reusable KV states. HeMeR tracks declarative, procedural, and working memory contents across action decisions, preserves KV states for unchanged contents, and reconciles only the KV states affected by memory-content updates. This enables efficient and stable inference as memory contents evolve during interaction. We evaluate HeMeR on ALFWorld, Habitat, and real-world robotic deployments requiring coordinated use of heterogeneous memory contents. In the real-world setting, HeMeR achieves 83.20% task success and reduces per-step inference time from 2.94 to 1.87 seconds compared with MemRL, while preserving compatible KV states as task progress updates the memory context.
Be the first to rate this article.

Comments & Questions (0)

Captcha
Please be respectful — let's keep the conversation friendly.

No comments yet

Be the first to comment!