PertMind：用细胞扰动数据的强化学习激发 LLM 涌现生物推理能力
PertMind: Eliciting Emergent Biological Reasoning in LLM via Reinforcement Learning on Cellular Perturbation Data
PertMind 将细胞扰动图谱用作强化学习环境，以实测基因响应作为可计算奖励，结合可信轨迹监督初始化与基因、通路、格式三级强化信号。该模型仅在正向扰动-响应预测上训练，却在未见细胞情境中提升响应推断，并无需针对性后训练即可迁移至反向扰动识别、双扰动推理、表型筛选优先级排序和生物过程解读。
Authors:Zhenchao Tang, Xiaogang Xu, Jiafei Wu, Jiahui Guan, Bo Li, Tianxu Lv, Jiale Zhou, Haohuai He, Zhi Song, Hanbo Huang, Jiehui Huang, Xun Lin, Zhipeng Deng, Zhaoxing Li, Guanxing Chen, Yaokun Li, Mengran Li, Songming Zhang, Zhe Liu
Abstract:Large language models can describe mechanisms, yet scalable post-training still depends on costly, manually curated biological reasoning traces. Here we show that cellular perturbation atlases can instead become reinforcement-learning environments, where measured gene responses provide computable rewards for biological reasoning. We introduce PertMind, which combines trusted-trajectory supervised initialization with gene-, pathway-, and format-level reinforcement signals. Although trained only on forward perturbation-response prediction, PertMind improves response inference in unseen cellular contexts while retaining general language capabilities. It also transfers, without task-specific post-training, to reverse perturbation identification, double-perturbation reasoning, phenotypic-screen prioritization, and biological-process interpretation. PertMind further generates biological profiles that support competitive gene, cell, and donor representations across multiscale downstream tasks. These results support the hypothesis that reinforcement on experimental endpoints can concentrate reusable biological strategies already accessible to pretrained models. More broadly, perturbation-derived reinforcement learning offers a scalable route for transforming expanding experimental atlases into training environments for general-purpose biological reasoning.
| Comments: | Project page: this https URL |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Quantitative Methods (q-bio.QM) |
| ACM classes: | I.2.6; I.2.7 |
| Cite as: | arXiv:2608.16419 [cs.LG] |
| (or arXiv:2608.16419v3 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2608.16419 arXiv-issued DOI via DataCite |
Submission history
From: Yaokun Li [view email]
[v1]
Mon, 17 Aug 2026 11:17:26 UTC (8,427 KB)
[v2]
Sat, 22 Aug 2026 05:21:26 UTC (1 KB) (withdrawn)
[v3]
Tue, 6 Oct 2026 15:42:49 UTC (8,366 KB)
来源：arXiv：cs.LG · arxiv.org