arXiv cs.AI· Pierre Llompart, Levent Guner, Helen Lai, Alessandro Tibo, Yijie Xu·· 6 天前AI 评分36
从 Benchmark 到实验台：AI 智能体能否胜任真实药物发现？
From Benchmark to Bench: Can Agents Survive Real-World Drug Discovery?
AI 导读
研究者开发了开源模块化智能体 MAGI，可自主设定目标、启动并监控分子优化、解读构效关系并调整策略，分子生成可直接由 LLM 完成或委托给 REINVENT 4。在来自三家药企的九个回顾性先导优化项目中，两条生成路线均产出有效结构，但项目能否达标取决于预测模型在候选化学空间上的准确度，而非生成路线。盲评中化学家无法区分智能体提案与留出化合物，认为其 SAR 推理大体合理但不完整。
正文 · 原文
Abstract:Agentic systems increasingly coordinate molecular-design tools, but it is unclear which layer of the stack limits outcomes on real projects. We developed MAGI, an open modular agent that authors objectives, launches and monitors optimization, interprets structure--activity relationships, and revises its strategy accordingly. MAGI generates molecules either directly through the LLM or by delegating to REINVENT 4, with scoring services interchangeable behind a common contract. We tested it across nine retrospective lead-optimization campaigns from three pharmaceutical companies, replayed under fixed temporal cutoffs. Both routes produced valid structures: LLM proposals stayed closer to local chemistry and reached comparable or higher primary activity in fewer operations, whereas REINVENT explored broader chemical space. Whether a campaign met its objective depended on the predictive models, not on the generation route: attainment followed model accuracy on the chemistry proposed, dropping once that chemistry moved outside the model's applicability domain. Separately, a blinded evaluation asked whether the MAGI's output could pass as expert work: chemists were not able to discriminate agentic proposals from held-out compounds, and judged the SAR reasoning broadly plausible yet incomplete. Together, these results position MAGI as a coordination layer pluggable into existing computational chemistry workflows. The ceiling on real projects, however, remains currently set by scorer applicability rather than by tool orchestration.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.06411 [cs.AI] |
| (or arXiv:2610.06411v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.06411 arXiv-issued DOI via DataCite |
Submission history
From: Pierre Llompart [view email]
[v1]
Mon, 5 Oct 2026 14:28:53 UTC (15,081 KB)
[v2]
Fri, 9 Oct 2026 08:05:52 UTC (14,424 KB)
来源：arXiv cs.AI · arxiv.org