Toward Controlling Biology with Language:Offline Learning of Prompt-Conditioned Interventions for Cells, Organoids, and Biobots
Abstract
Artificial intelligence increasingly serves as a natural-language interface to complex technical systems, letting people accomplish sophisticated tasks by describing what they want rather than specifying how to do it. Extending this interface to living systems is harder: unlike code or images, a biological intervention has no closed-form linguistic meaning, and the paired language-intervention-outcome data needed to learn such a mapping is expensive to collect, since each example requires its own wet-lab experiment. One way around this is to treat an existing archive of interventions and their already-observed outcomes as a fixed, offline dataset, and use a vision-language model to judge, without any new experiments, whether an archived outcome matches a natural-language description. But whether that judgment is reliable enough to train a language-to-intervention mapping on -- without new experiments and without human validation -- has remained unclear. Here we show that a natural-language interface for a living organism -- a xenobot, a synthetic multicellular construct with no nervous system -- can be learned entirely offline this way, using a vision-language model's own judgment as the sole training reward: an instruction is mapped to the intervention already on record as producing the described behavior. This mapping generalizes to entirely new instructions, evaluated against archive data withheld from training (80.0% held-out accuracy vs a $66.7% chance baseline, matching a network trained directly on ground-truth labels).
Related
- Region-Aware Degradation-Conditioned Vision-Language Learning for Blind Image-Quality Assessment of Power Field-Operation Imagery
- Protein Language Model-Conditioned Graph Neural Networks for Multitask GPCR Ligand Activity Prediction
- MIEN: a deep-learning model for pocket-conditioned glycan pose and conformation discrimination
- Reformulating Protein Corona Prediction as Universal Protein-conditioned Learning
- Pocket contrastive learning and conformation fusion for robust structure-conditioned 3D molecular generation under receptor conformational variation.
BibTeX
@misc{le2026toward,
title = {{Toward Controlling Biology with Language:Offline Learning of Prompt-Conditioned Interventions for Cells, Organoids, and Biobots}},
author = {Nam H. Le and Douglas Blackiston and Michael Levin and Josh Bongard},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.02247},
eprint = {2610.02247},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.02247},
note = {CC0},
}
RIS
TY - UNPB
TI - Toward Controlling Biology with Language:Offline Learning of Prompt-Conditioned Interventions for Cells, Organoids, and Biobots
AU - Nam H. Le
AU - Douglas Blackiston
AU - Michael Levin
AU - Josh Bongard
PY - 2026
DA - 2026/10/01
PB - arXiv
DO - 10.48550/arxiv.2610.02247
UR - https://arxiv.org/abs/2610.02247
AB - Artificial intelligence increasingly serves as a natural-language interface to complex technical systems, letting people accomplish sophisticated tasks by describing what they want rather than specifying how to do it. Extending this interface to living systems is harder: unlike code or images, a biological intervention has no closed-form linguistic meaning, and the paired language-intervention-outcome data needed to learn such a mapping is expensive to collect, since each example requires its own wet-lab experiment. One way around this is to treat an existing archive of interventions and their already-observed outcomes as a fixed, offline dataset, and use a vision-language model to judge, without any new experiments, whether an archived outcome matches a natural-language description. But whether that judgment is reliable enough to train a language-to-intervention mapping on -- without new experiments and without human validation -- has remained unclear. Here we show that a natural-language interface for a living organism -- a xenobot, a synthetic multicellular construct with no nervous system -- can be learned entirely offline this way, using a vision-language model's own judgment as the sole training reward: an instruction is mapped to the intervention already on record as producing the described behavior. This mapping generalizes to entirely new instructions, evaluated against archive data withheld from training (80.0% held-out accuracy vs a $66.7% chance baseline, matching a network trained directly on ground-truth labels).
ER -