ADEMA:一种用于长周期知识综合的知觉状态 orchestration 架构与 LLMAgents
ADEMA: A Knowledge-State Orchestration Architecture for Long-Horizon Knowledge Synthesis with LLMAgents
AI总结:
本文提出 ADEMA 架构,通过显式知识状态管理、双评估治理、自适应任务模式切换等方法,解决长周期知识综合中知识状态漂移、中间承诺隐含等问题,强调显式证据链和可恢复连续性。
摘要:
长周期 LLM 任务失败并非因为单个答案不可达,而是由于知识状态在轮次间漂移,中间承诺隐含,中断破坏演进的证据链。本文提出 ADEMA 作为长周期知识综合的知觉状态 orchestration 架构,而非通用多智能体运行时架构。该架构结合显式认知书目、异质双评估治理、自适应任务模式切换、声誉驱动的资源分配、可检查点恢复的持久性、分段记忆压缩、以 artifact 为先的组装,以及最终有效性检查与安全回退。证据完全来自现有材料:四个场景展示包、固定 60 次运行机制矩阵、针对性微消融和 artifact 链补充,以及修复的协议级基准,在其中代码导向的评估是最清晰的质量敏感机制块。在固定矩阵中,移除检查点/恢复导致唯一无效运行,且在中断敏感的恢复条件下。相比之下,双评估、分段合成和动态治理应被解释为支持控制机制,塑造轨迹纪律、显式 artifact 进展和成本-质量行为,而非通用二元前提。贡献是显式知识状态 orchestration 架构,其中显式认知状态转移、证据承载的 artifact 进展和可恢复连续性是主要设计承诺。
英文摘要:
Long-horizon LLM tasks often fail not because a single answer is unattainable, but because knowledge states drift across rounds, intermediate commitments remain implicit, and interruption fractures the evolving evidence chain. This paper presents ADEMA as a knowledge-state orchestration architecture for long-horizon knowledge synthesis rather than as a generic multi-agent runtime. The architecture combines explicit epistemic bookkeeping, heterogeneous dual-evaluator governance, adaptive task-mode switching, reputation-shaped resource allocation, checkpoint-resumable persistence, segment-level memory condensation, artifact-first assembly, and final-validity checking with safe fallback. Evidence is drawn entirely from existing materials: a four-scenario showcase package, a fixed 60-run mechanism matrix, targeted micro-ablation and artifact-chain supplements, and a repaired protocol-level benchmark in which code-oriented evaluation is the clearest quality-sensitive mechanism block. Across the fixed matrix, removing checkpoint/resume produced the only invalid run, and it did so in the interruption-sensitive resume condition. By contrast, dual evaluation, segment synthesis, and dynamic governance are best interpreted as supporting control mechanisms that shape trajectory discipline, explicit artifact progression, and cost-quality behavior rather than as universal binary prerequisites for completion. The contribution is therefore a knowledge-state orchestration architecture in which explicit epistemic state transition, evidence-bearing artifact progression, and recoverable continuity are the primary design commitments.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。