MaSRead：复制潜在存储的内容寻址读取
MaSRead: Content-Addressed Reading of Replicated Latent Stores
- INESC TEC
- ESTG, Instituto Politécnico de Viana do Castelo（维亚纳堡理工学院）
- FCUP, Universidade do Porto（波尔图大学理学院）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
MaSRead 是一种针对复制潜在存储的内容寻址读取方法，通过键控标签集路由与硬注意力掩码解码，可在各类存储中选择性恢复所需片段，且具备鲁棒性与模型迁移性。
AI中文摘要：
在潜在空间中推理的独立智能体可将计算状态作为键值缓存片段而非文本共享。这些片段通过无冲突复制数据类型（CRDT）合并，形成在任何传递顺序或复制下收敛的存储。但编码时未知的后续查询无法可靠读取合并后的缓存：共置片段会相互干扰，因此共置并非可寻址性。MaSRead 解决了对内容的读取问题，它通过从片段词派生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片段。在词汇连通性下，图遍历可到达多跳查询所需的片段。在链式、流水线式、对称式、中心式及自然语言存储中，MaSRead 能单独恢复已访问的片段，在无关片段累积时仍保持有效，且可迁移至另一模型家族。路由后，物化解码依赖于片段长度而非总存储大小；端到端工作仍包含依赖存储的路由及每个已访问片段的一次读取。其局限性明确：词汇路由可能遗漏不连通的证据，答案合成仍受限于冻结的阅读器。因此，当所需片段通过内容与查询相连时，复制潜在存储可对后续查询实现选择性读取。
英文摘要：
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. 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. 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. 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.
