从事件到推理:基于显式状态的层级规则组合

arXiv cs.AI 论文

摘要

论文提出事件推理系统(ERS),将推理状态显式表示为事件图,通过低层可复用规则与可递归组合的高层控制规则构建可解释、可修订的推理路径,在结构化事件图上实现长依赖链、零无关操作与高效的动态子目标控制。

arXiv:2610.22403v1 Announce Type: new Abstract: What a reasoning system emits is a conclusion; what it owes is a justification. Producing one requires every operation to be admissible on the facts actually available, the dependencies of intermediate results to survive to the end of the path, and the path to be revised when a premise changes. We present the Event Reasoning System (ERS), in which the reasoning state is an explicit event graph. Semantic features nominate candidate objects and rule bindings, and declared rule and task conditions decide which may execute, so candidates that violate the represented period, unit, or evidence constraints cannot enter an accepted reasoning path. Low-level rules provide reusable local transformations, and recursively composable high-level control rules organize subgoals and legal paths under the current goal, information gaps, and budget; a finite library is thereby rebound into paths never stored as complete solutions. We formalize admissible transitions, task-dependent control, compositional expressivity, dependency-preserving revision, and the gap between reachable solutions and those a budgeted controller finds. On structured-event graphs, ERS builds dependency chains of up to twelve steps, cuts irrelevant executed domain operations from 32 to 0 as lookahead grows to 16, and produces no unsupported target output across 24 negative cases; with rules frozen, it passes 12/12 focused checks comprising nine accepted derivations involving unseen input/path combinations and three justified refusals, as well as 6/6 fact-change cases. On fourteen jointly solved development tasks, the recorded dynamic subgoal configuration uses 51.6% less inference wall-clock time than single-level global control under their respective frozen settings.
查看原文
查看缓存全文

缓存时间: 2026/10/19 07:01

# From Events to Reasoning: Hierarchical Rule Composition over Explicit States
Source: [https://arxiv.org/abs/2610.22403](https://arxiv.org/abs/2610.22403)
[View PDF](https://arxiv.org/pdf/2610.22403)

> Abstract:What a reasoning system emits is a conclusion; what it owes is a justification\. Producing one requires every operation to be admissible on the facts actually available, the dependencies of intermediate results to survive to the end of the path, and the path to be revised when a premise changes\. We present the Event Reasoning System \(ERS\), in which the reasoning state is an explicit event graph\. Semantic features nominate candidate objects and rule bindings, and declared rule and task conditions decide which may execute, so candidates that violate the represented period, unit, or evidence constraints cannot enter an accepted reasoning path\. Low\-level rules provide reusable local transformations, and recursively composable high\-level control rules organize subgoals and legal paths under the current goal, information gaps, and budget; a finite library is thereby rebound into paths never stored as complete solutions\. We formalize admissible transitions, task\-dependent control, compositional expressivity, dependency\-preserving revision, and the gap between reachable solutions and those a budgeted controller finds\. On structured\-event graphs, ERS builds dependency chains of up to twelve steps, cuts irrelevant executed domain operations from 32 to 0 as lookahead grows to 16, and produces no unsupported target output across 24 negative cases; with rules frozen, it passes 12/12 focused checks comprising nine accepted derivations involving unseen input/path combinations and three justified refusals, as well as 6/6 fact\-change cases\. On fourteen jointly solved development tasks, the recorded dynamic subgoal configuration uses 51\.6% less inference wall\-clock time than single\-level global control under their respective frozen settings\.

## Submission history

From: Cheng Lyu \[[view email](https://arxiv.org/show-email/a6f8c8fd/2610.22403)\] **\[v1\]**Wed, 14 Oct 2026 12:47:40 UTC \(4,219 KB\)

相似文章

确定性事件图基质作为反事实推理的世界模型

arXiv cs.AI

该论文提出并评估了一类称为事件图基质的因果推理世界模型,通过确定性重放在类型化RDF事件日志上进行反事实查询,在多个基准上优于基线模型,同时保证了可检查性和可重放一致性。

抽象事件因果规则:归纳与应用

arXiv cs.AI

本文提出了抽象事件因果规则(AECR),一种关系级因果抽象范式,将具体的因果关系对转化为广义的因果逻辑。它引入了一个用于因果归纳的多智能体系统和一个基于注意力的编码器,从而改善事件预测,尤其是对稀有和未见事件。

递归推理系统的状态表示与终止条件

arXiv cs.AI

本文提出了一种用于递归推理系统的认知状态图表示方法以及一种基于顺序间隙的终止标准,旨在解决如何管理不断演变的推理状态以及何时停止迭代的问题。

ARBOR:通过推理状态跟踪定位长思维链中的推理失败

arXiv cs.CL

ARBOR 提出一种推理失败归因框架,将推理轨迹表示为推理状态树,以全局视角定位长思维链中的过度思考、思考不足和无依据推断等失败模式。研究发现该树状状态能区分真正导致失败的步骤与风险步骤。