model-evaluation 4 Numbers Eat Pipelines: Two Projects, One Habit of Distrusting Your Own Metrics Jun 22, 2026 How to Actually Report FID: A Single-GPU Reproducibility Checklist Jun 22, 2026 [Troubleshooting] Your FID of 0.24 Isn't Near-Perfect — It's the Wrong Feature Space Jun 18, 2026 Making an ESM2 Protein Variant Classifier Practical Jun 14, 2026