让世界模型可审计,确保预测状态与现实一致。
TRW: TRACE-RealWorld---An Auditable Consistency Contract for World Models as Materialized Views

- 将预测状态视为物化视图,用可过期的读取授权保证一致性
- 自适应刷新减少过时执行,局部修复降低9.56单位修复工作量
- 适合需要可审计预测的高风险场景如搜救系统
世界模型使智能体基于预测物理状态进行规划,但预测状态会漂移;重新观测成本高且延迟,修复可能失败。本文提出TRACE-RealWorld(TRW),据我们所知首个承诺级别的世界模型一致性合约。TRW将预测状态视为物化视图,将物理承诺视为可过期的读取授权。类型化、校准后的声明定义了后果条件下的新鲜度和验证价格。自适应刷新将双卡尔曼同步推广为在证据可能改变决策时查询世界;依赖范围内的SagaLLM补偿修复在授权后失效的可逆承诺。在事件对齐的风险预言机和恢复存活性假设下,我们证明单独使用同步或补偿不足,而其组合可提供条件一致性保证。否则,该框架可将违规分解为可命名的债务。我们在Flood-SAR(基于真实地理的搜救模拟器)中实现TRW,对六个预注册问题在冻结操作点上使用保留种子测试。自适应刷新减少过时执行,但未在成本、覆盖范围或救援结果上超越固定刷新。局部修复使每任务修复工作量减少9.56单位,恢复延迟缩短80.7秒;观察到的残差违规差异为零,未建立等价性。检测覆盖率为0.83–0.89,97次调用的恢复中有10次在任务结束时尚未完成。实验将经验松弛作为敏感性输入,而非违背定理假设或提供同时的人口证书。精确回放可重构有争议的调度。因此,TRW使世界模型成为可审计的预测接口,而非自我验证的真理来源。
原文摘要 · Abstract (English)
World models let agents plan against predicted physical state, but that state drifts; re-observation is costly and delayed, and repair can fail. We present TRACE-RealWorld (TRW), to our knowledge the first commitment-level consistency contract for world models. TRW treats predicted state as a materialized view and a physical commitment as a read whose authorization can expire. Typed, calibrated claims specify consequence-conditioned freshness and priced verification. Adaptive refresh generalizes dual-Kalman synchronization to consult the world when evidence could change a decision; dependency-scoped SagaLLM compensation repairs reversible commitments invalidated after authorization. Under an event-aligned risk oracle and recovery-liveness assumptions, we prove that synchronization and compensation are insufficient alone, whereas their composition gives a conditional consistency guarantee. Otherwise, the argument yields an auditable decomposition of violations into named debts. We implement TRW in Flood-SAR, a search-and-rescue simulator over real geography, and test six preregistered questions at frozen operating points on held-out seeds. Adaptive refresh reduces stale execution but does not dominate fixed refresh on cost, coverage, or rescue outcomes. Localized repair reduces repair work by 9.56 units per mission and restoration latency by 80.7 seconds relative to global recovery; the observed residual-violation difference is zero without establishing equivalence. Detection coverage is 0.83-0.89, and 10 of 97 invoked restorations are incomplete by mission end. The campaign treats empirical slack as sensitivity inputs rather than discharging the theorem's assumptions or providing a simultaneous population certificate. Exact replay reconstructs a disputed dispatch. TRW thus makes a world model an auditable predictive interface rather than a self-validating source of truth.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品;中文卡片由大模型生成,请以原文为准。