Abstract
AI agents operating over extended time horizons accumulate experiences that serve multiple concurrent goals, and must often maintain conflicting interpretations of the same events. A concession during a client negotiation encodes as a ``trust-building investment'' for one strategic goal and a ``contractual liability'' for another. Current memory architectures assume a single correct encoding, or at best support multiple views over unified storage. We propose Rashomon Memory: an architecture where parallel goal-conditioned agents encode experiences according to their priorities and negotiate at query time through argumentation. Each perspective maintains its own ontology and knowledge graph. At retrieval, perspectives propose interpretations, critique each other's proposals using asymmetric domain knowledge, and Dung's argumentation semantics determines which proposals survive. The resulting attack graph is itself an explanation: it records which interpretation was selected, which alternatives were considered, and on what grounds they were rejected. We present a proof-of-concept showing that retrieval modes (selection, composition, conflict surfacing) emerge from attack graph topology, and that the conflict surfacing mode, where the system reports genuine disagreement rather than forcing resolution, lets decision-makers see the underlying interpretive conflict directly.
Abstract (translated)
在长时间运行中，AI代理会积累服务于多个并发目标的经验，且往往需要对同一事件保持相互冲突的解释。例如，在客户谈判中做出的让步，对一个战略目标而言编码为“信任建设投资”，而对另一个目标则编码为“合同责任”。现有记忆架构通常假设存在唯一正确的编码方式，或最多支持在统一存储上提供多重视角。我们提出“罗生门记忆”（Rashomon Memory）：一种并行目标条件代理根据各自优先级编码经验，并在查询时通过辩论进行协商的架构。每个视角维护各自的本体论和知识图。检索时，各视角提出解释提案，利用不对称领域知识相互批评，并通过邓格的辩论语义决定哪些提案得以保留。由此产生的攻击图本身即是一种解释：它记录了所选解释、所考虑的替代方案，以及拒绝它们的依据。我们展示的概念验证表明，检索模式（选择、组合、冲突浮现）可从攻击图拓扑中自然涌现，且冲突浮现模式——即系统报告真实分歧而非强制解决——能让决策者直接洞察底层的解释冲突。
URL
https://arxiv.org/abs/2604.03588
