Study finds low-rank decoder bottleneck limits reliability of perturbation-prediction foundation models
A new arXiv paper argues that foundation models predicting cellular perturbation responses are judged mainly on aggregate accuracy, while a low-rank decoder bottleneck undermines their reliability for prioritizing experiments.
Read the full story at arXiv stat.ML ↗
Timeline · 1 report
- 2026-11-05 05:00 · arXiv stat.ML
A low-rank decoder bottleneck bounds reliability in foundation-model perturbation prediction