arXiv cs.AI· Tian Cai, Andrew Kleist, Jaimin Patel, Duccio Malinverni, Alexander K. Taylor, Malvika Sharan, Renato Umeton, M. Madan Babu·· 1 天前AI 评分32
AI for Biologists：智能体时代生物基础模型的知识边界
AI for Biologists: The Knowledge Boundaries of Foundation Models in the Age of Agentic Systems
AI 导读
一篇综述提出用数据、架构、训练方案、评估方案和分析后处理五个设计要素来界定生物基础模型的“知识边界”，即其输出可靠的输入、任务与条件范围。文章梳理了基于数百万核苷酸、蛋白质和单细胞测量训练的基础模型在预测蛋白质结构、突变效应和细胞响应上的成功与失败证据，并给出模型局限的标准化描述，以便研究者和替其执行任务的智能体可靠使用这些工具。
正文 · 原文
Abstract:Artificial intelligence (AI) is changing how biological research is done. Biological foundation models trained on millions of nucleotides, protein, and single-cell measurements predict protein structures, mutation effects, and cellular responses to genetic perturbations. Agentic systems, AI assistants built on large language models, can run multistep analyses from plain-language requests and increasingly call foundation models as tools. Yet it is often unclear when these models' outputs can be trusted. We frame this as a model's knowledge boundary: the range of inputs, tasks, and conditions over which its outputs are reliable. This review explains how foundation models are built and proposes that five design elements (data, architecture, training scheme, evaluation scheme, and post-analysis) define this boundary. We examine evidence on where current models succeed or fail and present standardized descriptions of model limits so that researchers, and the agents acting for them, can use these tools reliably.
| Comments: | To be published on Annual Review Biomedical Data Science |
| Subjects: | Biomolecules (q-bio.BM); Artificial Intelligence (cs.AI) |
| MSC classes: | 92C40, 68T07, 92-08 |
| ACM classes: | I.2.6; J.3 |
| Cite as: | arXiv:2611.17015 [q-bio.BM] |
| (or arXiv:2611.17015v1 [q-bio.BM] for this version) | |
| https://doi.org/10.48550/arXiv.2611.17015 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Tian Cai [view email]
[v1]
Thu, 12 Nov 2026 16:08:33 UTC (1,907 KB)
来源：arXiv cs.AI · arxiv.org