AI Paper Atlas

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

Adaptive Contextual Perception: How To Generalize To New Backgrounds and Ambiguous Objects

Zhuofan Ying, Peter Hase, Mohit Bansal

NeurIPS 2023 · Label-efficient and robust learning › OOD and anomaly detection · approach: Robust fine-tuning for OOD generalization

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
rotate foreground subspace into background subspace (reduce factorization) (ablation)ResNet features · SceneObjectfactorization (geometric): drop by 85.63\%Hurt
same rotation intervention (ablation)ResNet features · SceneObjectBackground-Invariance accuracy: drop by 18.24\%Hurt
same rotation intervention (ablation)ResNet features · SceneObjectObject-Disambiguation accuracy: increases by 5.97\%Improved
decrease foreground feature weighting by factor 64 (ablation)linear probe on features · SceneObjectBackground-Invariance accuracy: decreases by 0.63 pointsHurt
decrease foreground feature weighting by factor 64 (ablation)linear probe on features · SceneObjectObject-Disambiguation accuracy: increases by 0.68 pointsImproved
decrease background feature weighting by factor 64 (ablation)linear probe on features · SceneObjectBackground-Invariance accuracy: increases by 0.35 pointsImproved
decrease background feature weighting by factor 64 (ablation)linear probe on features · SceneObjectObject-Disambiguation accuracy: decreases by 0.92 pointsHurt
use proposed factorization/feature-weighting augmentation instead of ERMResNet · SceneObjectObject-Disambiguation accuracy: 71.93±1.27 -> 73.86±1.11Improved
use proposed augmentation instead of best baseline FishResNet · SceneObjectObject-Disambiguation accuracy: 71.30±0.74 -> 73.86±1.11Improved
use proposed augmentation instead of ERMResNet · ColorObjectBackground-Invariance accuracy: 29.11±3.39 -> 42.97±3.02Improved
use proposed augmentation instead of ERMResNet · ColorObjectObject-Disambiguation accuracy: 95.24±0.83 -> 92.02±1.06Hurt
use proposed augmentation instead of ERMResNet · ColorObjectfactorization: 71.27±1.14 -> 76.93±1.33Improved

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