让AI在有限记忆下持续保持长期推理能力
StatePlane: A Cognitive State Plane for Long-Horizon AI Systems Under Bounded Context
- 构建认知状态平面，动态管理记忆的生成与衰减
- 在六项任务中实现长程推理，无需扩展上下文窗口
- 适合需要长期交互的智能系统，如客服、助手
大语言模型（LLMs）和小语言模型（SLMs）受限于上下文窗口和键值（KV）缓存，难以在长时间交互中保持连贯推理。现有方法如扩展上下文、检索增强、摘要或静态文档，将记忆视为静态存储，无法在多轮、跨会话任务中保留决策相关状态。我们提出StatePlane，一种模型无关的认知状态平面，用于管理在有限上下文条件下运行的AI系统的事件、语义和程序状态的形成、演化、检索与衰减。基于认知心理学与系统设计，StatePlane实现了事件分割、信息论约束下的选择性编码、目标导向的检索与意图路由、重建式状态合成及自适应遗忘。本文提出形式化状态模型、感知KV的算法、安全治理机制（包括写入路径防污染）、企业集成路径及包含六个领域基准的评估框架。实验表明，长时序智能可在不扩大上下文窗口或重训练模型的前提下实现。
原文摘要 · Abstract (English)
Large language models (LLMs) and small language models (SLMs) operate under strict context window and key-value (KV) cache constraints, fundamentally limiting their ability to reason coherently over long interaction horizons. Existing approaches -- extended context windows, retrieval-augmented generation, summarization, or static documentation -- treat memory as static storage and fail to preserve decision-relevant state under long-running, multi-session tasks. We introduce StatePlane, a model-agnostic cognitive state plane that governs the formation, evolution, retrieval, and decay of episodic, semantic, and procedural state for AI systems operating under bounded context. Grounded in cognitive psychology and systems design, StatePlane formalizes episodic segmentation, selective encoding via information-theoretic constraints, goal-conditioned retrieval with intent routing, reconstructive state synthesis, and adaptive forgetting. We present a formal state model, KV-aware algorithms, security and governance mechanisms including write-path anti-poisoning, enterprise integration pathways, and an evaluation framework with six domain-specific benchmarks. StatePlane demonstrates that long-horizon intelligence can be achieved without expanding context windows or retraining models.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。