Untrained Neural Networks Hit 86.4% AUROC on ADBench Anomaly Detection
ideal-lab-sejong · hf · 2026-11-20
A paper from ideal-lab (Sejong University) proposes PRISM, which uses fixed random nonlinear features plus Mahalanobis scoring for anomaly detection — no weight optimization or pretrained representations.
- Tabular: achieves the highest mean AUROC (86.4%) and second-highest mean AUPRC (65.9%) across 47 ADBench datasets.
- Visual: on MVTec AD and VisA (27 categories), mean image-level and pixel-level AUROCs of 90.4% and 96.4%.
- Claim: anomaly separability may be intrinsic to data, questioning the assumption that learned representations are essential; theoretical analysis of nonlinear transformations included.
More from Research
- Transformers PR streams MoE experts from SSD to run a 180B model on a 32GB Mac — jomic01 · 2026-11-20
- AlphaFold3 and ESMFold2 fail on orphan proteins; Claude contact prediction still weak — sokrypton · 2026-11-20
- The Hundred-Page Machine Learning Book on sale for $14 (from $40) on Leanpub — burkov · 2026-11-20
- Image Models Now Ship Fine-tuned LLM Prompt Upsamplers — Why Not Fuse Them? — ostrisai · 2026-11-20
- PyTorch's ezyang on formalizing a trivial in-place kernel under PTX semantics — ezyang · 2026-11-20
- Scientist: Gemini 4 Neon designed innovative CAR receptors to kill cancer cells — DeryaTR_ · 2026-11-20