Cho-Jui Hsieh

Synthesizing Mechanistic Hypotheses from Single-Cell Omics via Discretized Feature Attribution and Empirical Language Model Grounding

This paper introduces a framework to convert continuous neural network attributions into testable biological hypotheses using decision trees and language models, validated on single-cell omics data.

Mechanistic Language Modeling and Oxygenated 3D Screening Reveal Berberine and Enzalutamide Synergy in Resistant Prostate Cancer