诚实仲裁问题：智能基础设施的认知拜占庭容错
The Honest Quorum Problem: Epistemic Byzantine Fault Tolerance for Agentic Infrastructure
浏览论文内容
中文总结 AI 辅助
研究智能基础设施中因推理错误导致的诚实仲裁问题，定义认知拜占庭容错（EBFT）模型，用两个量增强传统拜占庭故障界限，推导相关仲裁阈值条件及校准方法，指出添加代理提高容错的条件。
中文摘要 AI 辅助
状态机复制（SMR）和拜占庭容错（BFT）共识能保证在一定数量的任意勾结故障参与者情况下达成一致。但这些保证依赖于集合外参与者正确执行协议转换语义。智能验证者存在较弱边界：认证、响应、无歧义且符合协议的推理参与者可能因推理错误认可语义无效转换，即认知故障，这一集体现象为诚实仲裁问题。这样的仲裁能通过普通检查却为无效转换形成证书，仅达成一致不能保证语义有效性或执行安全性。此外，智能验证者常共享模型权重等，易受相关认知故障影响。我们定义了认知拜占庭容错（EBFT），一种用于智能基础设施和后确定性分布式系统的容错模型。EBFT用两个单独的、由置信索引的量增强传统拜占庭故障界限，分别界定拜占庭集合外的连贯无效认可和降低活性的不可用验证者支持。我们推导了语义有效性、共识一致、活性和可行阈值选择的仲裁阈值条件，并概述了估计这些预算的校准方法。我们表明，仅当添加名义上不同的代理可显著降低无效认可或不可用支持的上尾集中度时，才会提高容错能力。
英文摘要
State machine replication (SMR) and Byzantine fault-tolerant (BFT) consensus guarantee agreement despite a bounded number of arbitrary, colluding faulty participants. However, these guarantees rely on participants outside this set correctly executing the protocol's transition semantics. Agentic validators expose a weaker boundary: an authenticated, responsive, non-equivocating, and protocol-compliant reasoning participant may still endorse a semantically invalid transition due to reasoning errors. We call this failure mode an epistemic fault, and the collective phenomenon the Honest Quorum Problem (where "honest" means protocol-compliant, not semantically correct). Such a quorum can satisfy ordinary checks while forming a certificate for an invalid transition. Thus, agreement alone does not guarantee semantic validity or execution safety. Furthermore, because agentic validators often share model weights, training distributions, prompts, or toolchains, they are highly susceptible to correlated epistemic faults. We define Epistemic Byzantine Fault Tolerance (EBFT), a fault-tolerance model for agentic infrastructure and post-deterministic distributed systems. EBFT augments the conventional Byzantine fault bound with two separate, confidence-indexed quantities: $e_δ$ bounds coherent invalid endorsements outside the Byzantine set, and $u_ε$ bounds unusable validator support that degrades liveness. These quantities characterize semantic safety risk and liveness degradation independently. We derive quorum-threshold conditions for semantic validity, consensus agreement, liveness, and feasible threshold selection, and outline a calibration methodology for estimating these budgets. We show that adding nominally distinct agents improves fault tolerance only when it measurably reduces the upper-tail concentration of invalid endorsements or unusable support.
