factual
Cell modeling mirrors LLM pretraining then RL fine-tuning
NewLimit's modeling structure parallels LLM training: a base model predicts how TF perturbations change every gene the cell expresses (analogous to pretraining a general world representation), built on protein foundation models, then separate prediction 'heads' impose value judgments like 'this transcriptome looks younger' (analogous to RL on a specific objective), and additional heads could target arbitrary desired cell states.
factualpending
Speaker
Jacob KimmelEvidence Quote
“you could have multiple heads on a language model. One of those for us is actually just predicting every gene the cell is expressing... Then we also have something more like value judgments.”
Created: 6/13/2026, 3:28:32 AM
Your notes
Loading notes...