AI 中文总结
研究针对大语言模型在长期多智能体计算建模中因无状态架构受限的问题，提出Ensemble QSP多智能体框架，其分层内存架构可保持上下文稳定，经测试能自主进行药代动力学 - 药效学模型选择，提升参数恢复能力，且架构与领域无关。
AI 中文摘要
大语言模型（LLMs）虽有卓越推理能力，但无状态架构限制其在需要多会话连续性和定量严谨性的长期研究工作流程中的部署。本文提出Ensemble QSP，这是一个多智能体框架，具有三层分层内存架构。通过限制每个状态类别并清除已完成工作，使注入的上下文在项目持续时间内保持有界且恒定。系统在领域专家首席研究员的指导下协调五个专业工作智能体，通过基于物理的检查表和结构化领域知识执行物理约束。综合基准测试表明，该系统能在无人工干预的情况下进行强大的自主药代动力学 - 药效学模型选择，在不同成本的LLMs中结果质量一致，相对于单智能体基线提高了PK参数恢复能力，且在同一任务的不同语言提示下模型选择稳定。跨广泛复杂度范围的基于生理的药代动力学（PBPK）模型的特征级消融表明，首席研究员智能体监督提高了调试效率，同时在各种条件下保持最终准确性。该架构在结构上与领域无关，添加新的科学领域仅需新的首席研究员智能体配置。
英文摘要
Large language models (LLMs) demonstrate remarkable reasoning capabilities, yet their stateless architecture fundamentally limits deployment in long-horizon research workflows requiring multi-session continuity and quantitative rigor. Here we present Ensemble QSP, a multi-agent framework featuring a three-layer hierarchical memory architecture that bounds injected context (median 301 tokens, max 4,050) by capping state categories and evicting completed work. This enables continuous autonomous operation without context degradation. The system orchestrates five specialist worker agents under domain-expert principal investigators (PIs), enforcing physical constraints through physics-based checklists and structured domain knowledge. Comprehensive benchmarking demonstrates autonomous pharmacokinetic-pharmacodynamic (PKPD) model selection, improved parameter recovery relative to single-agent baselines, and robust interpretation of linguistically diverse prompts. Replication with open-weight models (DeepSeek-V4-Flash/Pro, Llama 3.1 70B) confirmed these architectural conclusions across PKPD modeling, literature synthesis, and PBPK model implementation, proving the framework is independent of proprietary LLMs. Feature-level ablations show that memory, retrieval, and PI oversight address distinct scientific failure modes, though underlying LLM capability remains consequential for stringent physical-consistency checks. The architecture is structurally agnostic to computational biology; adding a new scientific domain requires only a new PI-agent configuration.
Comments58 pages, 27 figures, 4 tables. Preprint submitted for publication
