Skip to content

PertMind: Eliciting Emergent Biological Reasoning in LLM via Reinforcement Learning on Cellular Perturbation Data

Aug 2026 · 0 citations · 55 references
Computer Science Biology

TL;DR

It is shown that cellular perturbation atlases can instead become reinforcement-learning environments, where measured gene responses provide computable rewards for biological reasoning, and introduces PertMind, which combines trusted-trajectory supervised initialization with gene-, pathway-, and format-level reinforcement signals.

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.

View source

Similar papers

CellRFT: Reinforcement Fine-Tuning for Single-Cell Perturbation Modeling

Comprehensive experiments demonstrate CellRFT's applicability across different pretrained models and effectiveness in improving perturbation prediction, reveal that optimizing one biological criterion can help or hinder others, and show that complementary rewards can improve criteria beyond those directly optimized.

Jie Yan, Li Liu, Han-Ze Guo et al. · 0 citations

EmbodiedMind: Adaptive Data Curation and Prefix-Tree Reinforcement Learning for Efficient Embodied Intelligence

Trie-GRPO, a novel reinforcement learning algorithm based on action prefix trees, which enables step-level advantage estimation, is introduced, which resolves the credit assignment problem by isolating intermediate correct decisions from downstream errors, while effectively balancing exploration efficiency and depth co...

Fei-Fan Wang, Zong-Bing Zhang, Yu Zhang et al. · 0 citations
Review Open access Sep 2026

Reinforcement Learning Post-Training for Reasoning Large Language Models: Methods, Systems, and Evaluation

This survey develops a mechanism-driven framework for separating effects of reinforcement learning, and organizes methods along three axes— learning-signal type, policy/data regime, and optimization granularity—and decomposes their training stacks into reusable motifs spanning advantage construction, drift control, den...

Liu Yang, Han Zhu, Zheng-Yang Zhong et al. · 1 citation
#artificial intelligence Preprint Oct 2026

VCLMU: Mechanism-Centric Virtual Cell World Modeling for Perturbation Response

A mechanism-centric virtual cell world model that represents cellular state as a set of Latent Mechanism Units (LMUs) and treats genetic perturbations as actions on these latent states supports mechanism-level latent state transition as a useful formulation for virtual cell models that aim to predict and interpret cell...

Yu-Wei Miao, Azim Dehghani Amirabad, S. Oloff et al. · 0 citations

Delayed Convergence and Emergent CoT Reliance in RL-Tuned Language Models

This analysis reveals that intermediate log-probability is an unreliable indicator for reasoning capability; instead, reasoning performance results from a shift of internal confidence allocation where RL fine-tuning delays internal convergence, exhibiting prolonged mid-layer exploration before converging sharply at the...

Pablo Pérez-Lázaro, Rocío Aznar-Gimeno, F. J. Lacueva-Pérez et al. · 0 citations
#natural language process... Preprint Sep 2026

ConsensusBench: Benchmark of Consensus Nodes for LLM Reasoning via Outcome Reward Densifying

This work develops a new reinforcement learning signal named ConsensusPR, which directly reduces the reward sparsity of outcome reward across long reasoning trajectories, and introduces ConsensusBench, a novel dataset designed to provide rule-based process-level signals.

Shi-Qi Yan, Chao-Hong Tan, Qian Chen et al. · 0 citations

Related blog posts

Microsoft Research Blog Nov 16, 2026

TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficient

AI coding agents are increasingly help developers fix bugs and build features largely on their own, but every interaction with their models consumes tokens. In each session, an agent sends instructions, tool descriptions, and session history to its model. Most efforts to reduce this cost focus on individual tasks. Yet developers return to the same […] The post TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficient appeared first on Microsoft Research.

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.