让AI在推理出错时精准修复，不重来全程。
ORDDAR: Observation-Driven Reasoning for Distortion-Resilient Decision, Action, and Cognitive Recovery
- 将推理看作认知状态转移，定位错误环节
- 从过往经验中检索并修复受影响状态
- 适用于需要可靠决策的复杂任务场景
AI智能体在执行长期推理、规划、工具使用和自主决策时，中间状态错误会传播并导致不一致结果。现有方法多依赖迭代规划、自我反思或增强记忆，但很少能精确定位并选择性修复故障推理。本文提出ORDDAR（观察驱动的失真鲁棒决策、行动与认知恢复）框架，将推理建模为认知状态转移，检测局部失真，从过往经验中检索相关推理，并仅修复受影响的状态。因此，ORDDAR在局部推理转移层面实现恢复，而非重新生成完整轨迹。在数学、常识、多跳及临床推理等多个基准上的实验表明，相比多个基线方法，ORDDAR在推理质量、恢复能力与可解释性方面均有提升。
原文摘要 · Abstract (English)
AI agents increasingly perform long-term reasoning, planning, tool use, memory integration, and autonomous decision making, yet erroneous intermediate states can propagate and cause inconsistent decisions and unreliable outputs. Existing reasoning approaches mainly rely on iterative planning, self-reflection, augmented memory, or verification, but rarely localize and selectively repair faulty reasoning. We present ORDDAR (Observation-Driven Reasoning for Distortion-Resilient Decision, Action, and Cognitive Recovery), a reasoning framework that models reasoning as cognitive state transitions, detects localized distortions, retrieves related reasoning from prior experiences, and repairs only the affected states. ORDDAR therefore performs recovery at the local reasoning-transition level rather than regenerating the complete trajectory. Experiments across mathematical, commonsense, multi-hop, and clinical reasoning benchmarks demonstrate improved reasoning quality, recovery ability, and interpretability over multiple evaluated reasoning baselines.
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。