arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2610.21165cs.CL
收藏

从长上下文到选择性记忆:面向大语言模型的以记忆为中心的压缩方法综述

From Long Context to Selective Memory:A Survey of Memory-Centric Compression for Large Language Models

USTB(北京科技大学)

Chaogui Gou

AI总结:

本综述将大语言模型长上下文压缩重新定义为推理资源分配的记忆管理问题,统一多类相关研究,指出核心挑战是决策信息参与推理的形式与时机,提出未来研究方向。

摘要:

长上下文能力已成为大语言模型(LLMs)处理复杂任务的核心要求,但持续扩展上下文窗口并未消除推理时产生的计算、存储及信息选择压力。现有关于提示词压缩、KV缓存管理、高效服务、检索增强及智能体记忆的研究通常被孤立讨论,导致对存储与计算间的信息流动缺乏充分理解。本综述将长上下文压缩重新定义为推理时资源分配的记忆管理问题,区分了直接参与模型计算的计算活跃记忆,以及处于计算路径之外但可存储、检索、更新或遗忘的计算非活跃记忆,并将卸载KV缓存、持久前缀缓存及预计算KV状态表征为激活路径上的中间状态。在此基础上,我们将写入、卸载、召回、检索与注入、物化、驱逐、整合及遗忘定义为信息生命周期内的状态转换,并引入预算成本模型将这些操作置于单一成本账户中。该框架统一了短期上下文压缩、长期记忆管理及二者间的转换,表明核心挑战不仅是减少信息总量,而是决定哪些信息应以何种形式、在何时参与高成本推理。我们还总结了评估维度、基准缺口及开放挑战,认为未来研究应从静态、逐点压缩转向动态、可管控、感知生命周期的记忆管理。

英文摘要:

Long-context capability has become essential for large language models (LLMs) to handle complex tasks, yet continuously expanding the context window does not remove the computational, storage, and information-selection pressures incurred at inference time. Existing work on prompt compression, KV-cache management, efficient serving, retrieval augmentation, and agent memory is usually discussed in isolation, leaving the flow of information between storage and computation insufficiently understood. This survey reframes long-context compression as a memory-management problem for allocating inference-time resources. We distinguish compute-active memory, which directly participates in model computation, from compute-inactive memory, which lies outside the computation path but can be stored, retrieved, updated, or forgotten, and we characterize offloaded KV caches, persistent prefix caches, and precomputed KV states as intermediate states on the activation path. On this basis, we formulate writing, offloading, recall, retrieval-and-injection, materialization, eviction, consolidation, and forgetting as state transitions within an information lifecycle, and introduce a budgeted cost model that places these operations in a single cost account. The resulting framework unifies short-term context compression, long-term memory management, and the transitions between them, and shows that the central challenge is not merely reducing information volume but deciding which information should participate in costly inference, when, and in what form. We further summarize evaluation dimensions, benchmark gaps, and open challenges, and argue that future research should move from static, pointwise compression toward dynamic, governable, lifecycle-aware memory management.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑