Accepted Papers

Spotlight Papers

• Knowledge graphs beat foundation models for gene perturbation prediction, but only the non-causal ones.
  Sebastien Boyer ⋅ AMIAUD--PLACHY Ilann ⋅ Oliver Bent
• Open-Weight MLLMs Cannot Compare What They Cannot Name
  Wenxin Su ⋅ Anna Kreshuk
• Treating the Symptom, Missing the Cause: Anatomy of a Scientific Reasoning Failure
  Anna Gogleva ⋅ Martin Balla ⋅ Thomas Charman ⋅ Karl Gemayel
• Causal variant underestimation is a major overlooked driver of sequence-to-function model underperformance
  Shiron Drusinsky ⋅ Katherine S Pollard
• Misalignments in Generative Protein Design: Thermodynamic Filtering Fails to Predict Specificity
  Safwan Diwan ⋅ Mattias Tolhurst ⋅ Jeff Nivala
• Sparse crosscoders identify antibody features gained during fine-tuning of protein language models
  Wiona Glänzer ⋅ Sai Reddy ⋅ Alexander Yermanos