物理科學中的深度研究：多智能體框架與全面基準

摘要

深度研究代理（Deep Research Agents）是以大型語言模型（LLM）為基礎的系統，專為自主、多步驟的科學推理而設計，在加速物理科學研究方面具有巨大潛力。然而，目前對其在該領域能力的全面深入評估仍有所欠缺。為填補此空白，我們提出 PhySciBench，一個與物理科學研究高度相關的基準，包含 200 道經專家策劃的問題，涵蓋物理與化學兩大學科，分布於反映實際科學工作流程的六個任務類別。在 PhySciBench 上對當前最先進模型與代理系統的評估顯示其表現有限；即使是最強的基線模型 Gemini Deep Research，準確率也僅達 33.5%。對失敗案例的分析揭示了三個反覆出現的缺陷：延展推理鏈的脆弱性、跨步驟知識遷移能力不足，以及缺乏基於物理學的自我驗證機制。基於這些發現，我們開發了 DelveAgent，一個模組化多代理框架，配備自適應規劃循環、雙粒度記憶，以及分層的物理學基礎反思機制。在四個科學基準測試中，DelveAgent 將準確率提升最多 7.5 個百分點，同時將推理成本降至最強基線模型約三分之一。這些結果確立了 PhySciBench 作為評估物理科學中 AI 系統的重要基準，同時證明架構專業化能有效提升自主科學研究的可靠性。

English

Deep research agents are Large Language Model (LLM)-based systems designed for autonomous, multi-step scientific reasoning, and they hold immense potential for accelerating research in the physical sciences. However, comprehensive and in-depth evaluations of their capabilities within this domain remain lacking. To address this gap, we introduce PhySciBench, a benchmark highly relevant to physical science research, comprising 200 expert-curated questions, balanced between physics and chemistry, across six task categories that reflect real-world scientific workflows. Evaluations of state-of-the-art models and agent systems on PhySciBench reveal limited performance; even the strongest baseline, Gemini Deep Research, achieves an accuracy of only 33.5%. Analysis of failure cases identifies three recurrent deficiencies: fragility in extended reasoning chains, limited knowledge transfer across steps, and a lack of physics-grounded self-verification. Motivated by these findings, we develop DelveAgent, a modular multi-agent framework equipped with an adaptive planning loop, dual-granularity memory, and a hierarchical physics-grounded reflection mechanism. Across four scientific benchmarks, DelveAgent improves accuracy by up to 7.5 percentage points while reducing inference costs to approximately one-third of the strongest baseline. These results establish the significance of PhySciBench as a critical benchmark for evaluating AI systems in the physical sciences and demonstrate that architectural specialization can effectively enhance the reliability of autonomous scientific research.