Stanford's Kundaje challenges EVO2: simpler models already handle enhancer design
anshulkundaje · x · 2026-10-30
Computational biology professor Anshul Kundaje pushes back on @garykbrixi's vision of using a long-context EVO2 generator plus a supervised model to design entire genomic domains for tuning gene expression. He praises genomic in-context learning (ICL) but argues ICL-based enhancer design already benefits greatly from cell-type-specific sequence-to-ontology models, and simpler, faster approaches can accomplish enhancer/promoter design without EVO2.
More from Research
- Stanford releases RT-J, a System 1 model predicting structured data in one forward pass — jure · 2026-10-30
- Signal65 Launches PINNACLE Coding Agent Benchmark: One Model Aces 97% of Samples, Fails All 30 Jobs — ryanshrout · 2026-10-30
- TEM: Transformer Energy-Based Models Claim to Outscale Diffusion Without Guidance — du_yilun · 2026-10-30
- VisualForce: robots infer grip force from soft gripper deformation, no force sensors needed — du_yilun · 2026-10-30
- Microsoft Research: coding agents' real bottleneck is understanding code, not writing it — Microsoft Research · 2026-10-30
- HALO (ICML 2026 Spotlight): certified-stable RL keeps humanoids coordinated with unpredictable humans — Scobleizer · 2026-10-30