ADEMA: A Knowledge-State Orchestration Architecture for Long-Horizon Knowledge Synthesis with LLMAgents
2026-04-28 • Artificial Intelligence
Artificial Intelligence
AI summaryⓘ
The authors discuss why long tasks using large language models (LLMs) often fail due to knowledge drifting and interruptions breaking the thinking process. They propose ADEMA, a system designed to carefully track knowledge changes, manage evidence, and allow tasks to pause and resume safely. Their experiments show that being able to checkpoint and resume is crucial for handling interruptions, while other features help keep the process organized but are not absolutely required. Overall, the authors present ADEMA as a way to keep the knowledge state clear and recoverable during long, complex tasks.
large language modelsknowledge statecheckpointingmulti-agent systemsevidence trackingtask resumptionstate orchestrationartifact progressionevaluation mechanismsgovernance
Authors
Zhou Hanlin, Chan Huah Yong
Abstract
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.