提出RES架构，让AI助手查询海量文献时成本恒定、不编造内容。
Reasoner-Executor-Synthesizer: Scalable Agentic Architecture with Static O(1) Context Window
- 分三层处理：先理解意图，再用统计摘要检索数据，最后生成回答。
- 无论查4万还是1630万篇论文，平均仅用1574个词元，成本不变。
- 不用大模型看原始数据，从机制上杜绝幻觉，适合学术研究场景。
将大型语言模型（LLMs）部署为自主智能体时，常采用检索增强生成（RAG）技术，将检索到的文档输入上下文窗口，导致两个问题：上下文越长，幻觉风险越高；词元成本随数据集规模线性增长。我们提出一种三层次的推理-执行-合成（RES）架构，严格分离意图解析（Reasoner）、确定性数据检索与聚合（Executor）以及叙事生成（Synthesizer）。Executor不使用任何大模型词元，仅向Synthesizer传递固定大小的统计摘要。我们形式化证明，RES在数据集规模上达到O(1)词元复杂度，并在基于Crossref API（含1300多万篇文章）构建的学术研究助手ScholarSearch上验证。在100次基准测试中，无论数据集包含4.2万或1630万篇文章，平均词元消耗均为1574个。该架构通过设计实现数据幻觉消除：大模型从未见过原始记录。
原文摘要 · Abstract (English)
Large Language Models (LLMs) deployed as autonomous agents commonly use Retrieval-Augmented Generation (RAG), feeding retrieved documents into the context window, which creates two problems: the risk of hallucination grows with context length, and token cost scales linearly with dataset size. We propose the Reasoner-Executor-Synthesizer (RES) architecture, a three-layer design that strictly separates intent parsing (Reasoner), deterministic data retrieval and aggregation (Executor), and narrative generation (Synthesizer). The Executor uses zero LLM tokens and passes only fixed-size statistical summaries to the Synthesizer. We formally prove that RES achieves O(1) token complexity with respect to dataset size, and validate this on ScholarSearch, a scholarly research assistant backed by the Crossref API (130M+ articles). Across 100 benchmark runs, RES achieves a mean token cost of 1,574 tokens regardless of whether the dataset contains 42,000 or 16.3 million articles. The architecture eliminates data hallucination by construction: the LLM never sees raw records. KEYWORDS LLM agents; agentic architecture; hallucination elimination; token optimization; context window; retrieval-augmented generation; deterministic execution; scholarly metadata; Crossref API; O(1) complexity.
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。