MaSRead: Content-Addressed Reading of Replicated Latent Stores

MaSRead：复制潜在存储的内容寻址读取

Abstract: Independent agents that reason in latent space can share computed state as key-value cache fragments rather than text. Merged by a conflict-free replicated data type, these fragments form a store that converges under any delivery order or duplication. Yet a later query, unknown at encode time, cannot reliably read the merged cache: colocated fragments interfere, so colocation is not addressability.

摘要： 在潜在空间中进行推理的独立智能体可以共享计算状态，将其作为键值缓存片段而非文本进行传输。通过无冲突复制数据类型（CRDT）合并后，这些片段形成了一个存储库，无论交付顺序如何或是否存在重复，该存储库都能实现收敛。然而，在编码时未知的后续查询无法可靠地读取合并后的缓存：共存的片段会产生干扰，因此“共存”并不等同于“可寻址性”。

MaSRead addresses the read to content. It routes through opaque keyed tag sets derived from fragment words and decodes each selected fragment under a hard attention mask that hides the rest. Under lexical connectivity, a graph walk reaches the fragments required by a multi-hop query.

MaSRead 实现了基于内容的读取。它通过从片段词汇中导出的不透明键控标签集进行路由，并在隐藏其余部分的硬注意力掩码下解码每个选定的片段。在词汇连通性的基础上，通过图遍历（graph walk）可以获取多跳查询所需的片段。

Across chain, pipeline, symmetric, hub, and natural-language stores, MaSRead recovers visited fragments in isolation, remains effective as unrelated fragments accumulate, and transfers to another model family. After routing, materialized decoding depends on fragment length rather than total store size; end-to-end work still includes store-dependent routing and one read per visited fragment.

在链式、流水线、对称、中心式和自然语言存储等多种架构中，MaSRead 能够独立恢复已访问的片段，在无关片段累积时依然保持有效，并可迁移至其他模型系列。路由完成后，具体化解码（materialized decoding）仅取决于片段长度而非存储库的总大小；端到端的计算工作仍包含依赖于存储库的路由以及对每个已访问片段的一次读取。

The limits are explicit: lexical routing can miss disconnected evidence, and answer composition remains bounded by the frozen reader. Thus a replicated latent store becomes selectively readable for later queries when the needed fragments connect to the query through content.

其局限性是明确的：词汇路由可能会遗漏不连通的证据，且答案的组合仍受限于固定的阅读器。因此，当所需的片段通过内容与查询建立连接时，复制的潜在存储库便可针对后续查询实现选择性读取。