Score Distributions, Not Cells: Evaluating Single-Cell Perturbations Under Class Overlap

arXiv:2607.04595 · cs.LG, q-bio.QM, stat.ML · Submitted 2026-07-06 · Read on arXiv

Youssef Marrakchi, Davide D'Ascenzo, Sebastiano Cultrera di Montesano

cs.LG, q-bio.QM, stat.ML

Submitted: 2026-07-06

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Accuracy — 34× in this paper · explained in 1 episode
• Transformer — 14× in this paper · explained in 2 episodes
• Virtual Cell Challenge — 11× in this paper
• Benchmark — 7× in this paper · explained in 2 episodes
• Classifier Discrimination Score — 5× in this paper
• Perturbation Discrimination Score — 4× in this paper
• Benchmarking — 2× in this paper · explained in 4 episodes
• Adam — 1× in this paper · explained in 3 episodes
• Clustering — 1× in this paper · explained in 2 episodes
• Commentary — 1× in this paper · explained in 2 episodes
• Expressivity — 1× in this paper · explained in 2 episodes
• Generative Model — 1× in this paper · explained in 2 episodes
• Transformer Encoder — 1× in this paper · explained in 2 episodes

Sources

• Effects of Distance Metrics and Scaling on the Perturbation Discrimination Score
• PerturBench: Benchmarking Machine Learning Models for Cellular Perturbation Analysis

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks