AI Paper Atlas

Research map / Label-efficient and robust learning / OOD and anomaly detection

MOOD: Benchmarking Post-Hoc OOD Detection for Materials Property Prediction

Shirley X Yu, Amey Pasarkar, Adji Bousso Dieng

NeurIPS 2026 · Label-efficient and robust learning › OOD and anomaly detection · approach: Spurious-aware OOD detection · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Introduces MOOD, a benchmark testing whether post-hoc out-of-distribution detectors built for image classification transfer to graph neural networks for materials property prediction.

Problem

Post-hoc OOD detection (flagging OOD inputs without retraining) has been validated almost only on image classification, yet is needed for high-stakes scientific regression such as materials property prediction with GNNs.

Approach

A benchmark that evaluates existing post-hoc OOD detectors on GNNs for materials property prediction, rather than proposing a new detector.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
apply vision post-hoc OOD detectors to frozen materials GNN embeddings15 post-hoc detectors · MOOD (264 settings)mean AUROC: best 0.531 (Cosine), chance 0.5No clear effect
use supervised linear probe instead of unsupervised detectors (ablation)frozen GNN embeddings · dielectricAUROC: 0.522 -> 0.599Improved
use supervised linear probe instead of unsupervised detectors (ablation)frozen GNN embeddings · log_gvrhAUROC: 0.524 -> 0.600Improved
use supervised linear probe instead of unsupervised detectors (ablation)frozen GNN embeddings · log_kvrh, perovskites, phononsAUROC: probe 0.494 / 0.500 / 0.499No clear effect
replace linear probe with MLP probe (ablation)frozen GNN embeddings · MOOD encoder-bottlenecked tasksprobe AUROC: mean |delta| = 0.023, r = 0.971No clear effect
restrict to settings where OOD error exceeds ID error (ablation)Cosine detector · MOOD degraded subset (N=153)OLS slope of AUROC on MAE gap: beta +0.322 -> +0.598; R2 0.27 -> 0.45Improved
use isolation or boundary detectors instead of distance/density (ablation)post-hoc OOD detectors · MOOD degraded subsetslope of AUROC vs MAE gap: IForest +0.185, ViM +0.230 vs Cosine +0.598Hurt
use distance-based family instead of other families (ablation)post-hoc OOD detectors · MOOD (264 settings)win rate: 40% vs 20% random baselineImproved
abstain on samples flagged by detector (risk-coverage)post-hoc detectors · MOOD seed 42 encodersselective prediction effectiveness: SPE 0.14 to 0.37 (oracle = 1)Mixed

Experimental setup

Benchmarks: MOOD

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy