arXiv cs.MA· Juan B. P\'erez-S\'anchez, Viviana Roman-Ventura, Olalla Nieto-Faza, Aiwei Yin, Al\'an Aspuru-Guzik, Varinia Bernales·· 11 小时前AI 评分31
El Agente Quntur：面向 CASSCF 的智能体活性空间选择
El Agente Quntur: Agentic Active-Space Selection for CASSCF
AI 导读
El Agente Quntur 多智能体框架以 GPT-5.6 Sol 和 GPT-5.6 Terra 为语言模型，在 32 个固定几何构型的单重态激发计算中自主选择并修正 CASSCF 活性空间，相比六种确定性 Active Space Finder 协议将平均绝对误差降低 48%。
正文 · 原文
Abstract:Multireference calculations require time-consuming expert decisions about active spaces, state identity, and convergence. This study asks whether agents can make these decisions reliably. Using the El Agente Quntur multi-agent harness with GPT-5.6 Sol and GPT-5.6 Terra as its language models, agents study 32 fixed-geometry singlet excitations, detect missing active orbitals, compare excited states, and revise state assignments as calculations progress. Prompts specify equal-weight two-singlet state-averaged complete active space self-consistent field (CASSCF) calculations followed by strongly contracted second-order N-electron valence state perturbation theory (NEVPT2) with def2-TZVPD; agents choose and refine the active spaces. Excitation energies are evaluated against published complete active space second-order perturbation theory (CASPT2) reference values at matching geometries. Agentic execution reduces mean absolute error by 48\% relative to six deterministic Active Space Finder protocols. Computational traces vary substantially across runs in selection strategies and cost, even for the same molecule. These results underscore the advantage of agentic active space selection over fixed workflows, yet leave room to improve efficiency through runtime-aware decision-making, reduced inter-agent communication overhead and redundancy, more systematic checks of state identity, and clearer stopping criteria.
| Subjects: | Chemical Physics (physics.chem-ph); Multiagent Systems (cs.MA) |
| Cite as: | arXiv:2610.38145 [physics.chem-ph] |
| (or arXiv:2610.38145v2 [physics.chem-ph] for this version) | |
| https://doi.org/10.48550/arXiv.2610.38145 arXiv-issued DOI via DataCite |
Submission history
From: Juan B. Pérez-Sánchez [view email]
[v1]
Wed, 28 Oct 2026 01:28:33 UTC (19,022 KB)
[v2]
Fri, 30 Oct 2026 02:16:57 UTC (25,016 KB)
来源：arXiv cs.MA · arxiv.org