A Multi-Stage Rule-Chaining Framework for Compositional and Interpretable Cognitive Reasoning

论文概要

研究领域: ML 作者: Deblina Kar 发布时间: 2026-09-11 arXiv: 2509.05826

中文摘要

抽象与推理语料库(ARC)用于评测认知泛化能力，即从有限示例中推断和应用抽象规则的能力。本文提出多阶段规则链框架，在符号、结构和概念层面执行组合推理。框架整合三个互补求解器：(1) 确定性规则发现模块，通过几何、颜色和对象分析归纳原子变换；(2) 模式组合引擎，通过块合并、重复和空间启发式重构输出；(3) 结构抽象层，推断网格间的层级和嵌套关系。这些求解器在渐进式回退层级中顺序运行，每个阶段复用先前推理轨迹以增强可解释性和泛化。训练通过1000项任务中的995项，进一步在120项中评估105项，并解决240项ARC-AGI-2测试任务中的230项。系统在确定性、组合和抽象类别上均表现强劲，整体准确率超过95%。所提架构桥接符号推理与模式合成，为认知泛化提供可解释的洞察。结果表明，规则链和层级组合可在不依赖任务特定调优的情况下，将机器推理推向透明、类人一致的抽象。

原文摘要

The Abstraction and Reasoning Corpus (ARC) benchmarks cognitive generalization, the ability to infer and apply abstract rules from limited examples. This paper presents a multi-stage rule-chaining framework that performs compositional reasoning across symbolic, structural, and conceptual levels. The framework integrates three complementary solvers: (1) a deterministic rule discovery module that induces atomic transformations through geometric, color, and object-based analysis; (2) a pattern-composition engine that reconstructs outputs via block merging, repetition, and spatial heuristics; and (3) a structural abstraction layer that infers hierarchical and nested relationships across grids. These solvers operate sequentially within a progressive fallback hierarchy, where each stage reuses p...

#论文 #arXiv #ML #小凯