When Accuracy Is Not Enough: Uncertainty Collapse between Noisy Label Learning and Out-of-Distribution Detection

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Peng, Ningkang, Mao, Jingyang, Zhou, Runhan, Ma, Peirong, Gu, Yanhui
Natura:	Preprint
Pubblicazione:	2026
Soggetti:	Machine Learning Computer Vision and Pattern Recognition
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

HamBR: Active Decision Boundary Restoration Based on Hamiltonian Dynamics for Learning with Noisy Labels
di: Peng, Ningkang, et al.
Pubblicazione: (2026)

Holistic Reliability Propagation: Decoupling Annotation and Prediction for Robust Noisy-Label
di: Mao, Jingyang, et al.
Pubblicazione: (2026)

GAMR: Geometric-Aware Manifold Regularization with Virtual Outlier Synthesis for Learning with Noisy Labels
di: Peng, Ningkang, et al.
Pubblicazione: (2026)

Radial-Angular Geometry for Reliable Update Diagnosis in Noisy-Label Learning
di: Peng, Ningkang, et al.
Pubblicazione: (2026)

Early High-Frequency Injection for Geometry-Sensitive OOD Detection
di: Cheng, Chuanjie, et al.
Pubblicazione: (2026)

Learning with Adaptive Prototype Manifolds for Out-of-Distribution Detection
di: Peng, Ningkang, et al.
Pubblicazione: (2026)

Is Complex Training Necessary for Long-Tailed OOD Detection? A Re-think from Feature Geometry
di: Peng, Ningkang, et al.
Pubblicazione: (2026)

VMF-GOS: Geometry-guided virtual Outlier Synthesis for Long-Tailed OOD Detection
di: Peng, Ningkang, et al.
Pubblicazione: (2026)

Breaking Semantic Hegemony: Decoupling Principal and Residual Subspaces for Generalized OOD Detection
di: Peng, Ningkang, et al.
Pubblicazione: (2026)

Halt the Hallucination: Decoupling Signal and Semantic OOD Detection Based on Cascaded Early Rejection
di: Peng, Ningkang, et al.
Pubblicazione: (2026)

When the Small-Loss Trick is Not Enough: Multi-Label Image Classification with Noisy Labels Applied to CCTV Sewer Inspections
di: Chelouche, Keryan, et al.
Pubblicazione: (2024)

Neural Collapse-Inspired Multi-Label Federated Learning under Label-Distribution Skew
di: Peng, Can, et al.
Pubblicazione: (2025)

Adaptive Label Smoothing for Out-of-Distribution Detection
di: Xu, Mingle, et al.
Pubblicazione: (2024)

Distributional Uncertainty for Out-of-Distribution Detection
di: Kim, JinYoung, et al.
Pubblicazione: (2025)

Uncertainty-Participation Context Consistency Learning for Semi-supervised Semantic Segmentation
di: Yin, Jianjian, et al.
Pubblicazione: (2024)

How to Achieve Prototypical Birth and Death for OOD Detection?
di: Peng, Ningkang, et al.
Pubblicazione: (2026)

Zero-Shot Out-of-Distribution Detection with Outlier Label Exposure
di: Ding, Choubo, et al.
Pubblicazione: (2024)

Noisy Label Refinement with Semantically Reliable Synthetic Images
di: Li, Yingxuan, et al.
Pubblicazione: (2025)

Pairwise Similarity Distribution Clustering for Noisy Label Learning
di: Bai, Sihan
Pubblicazione: (2024)

Pursuing Feature Separation based on Neural Collapse for Out-of-Distribution Detection
di: Wu, Yingwen, et al.
Pubblicazione: (2024)

PLReMix: Combating Noisy Labels with Pseudo-Label Relaxed Contrastive Representation Learning
di: Liu, Xiaoyu, et al.
Pubblicazione: (2024)

Network Inversion for Uncertainty-Aware Out-of-Distribution Detection
di: Suhail, Pirzada, et al.
Pubblicazione: (2025)

Accuracy is Not Enough: Poisoning Interpretability in Federated Learning via Color Skew
di: Riya, Farhin Farhad, et al.
Pubblicazione: (2025)

Leveraging Perturbation Robustness to Enhance Out-of-Distribution Detection
di: Chen, Wenxi, et al.
Pubblicazione: (2025)

Uncertainty-Aware Likelihood Ratio Estimation for Pixel-Wise Out-of-Distribution Detection
di: Hölle, Marc, et al.
Pubblicazione: (2025)

Uncertainty-Guided Appearance-Motion Association Network for Out-of-Distribution Action Detection
di: Fang, Xiang, et al.
Pubblicazione: (2024)

Affordance-First Decomposition for Continual Learning in Video-Language Understanding
di: Xu, Mengzhu, et al.
Pubblicazione: (2025)

Estimating Noisy Class Posterior with Part-level Labels for Noisy Label Learning
di: Zhao, Rui, et al.
Pubblicazione: (2024)

AEON: Adaptive Estimation of Instance-Dependent In-Distribution and Out-of-Distribution Label Noise for Robust Learning
di: Garg, Arpit, et al.
Pubblicazione: (2025)

Learning Camouflaged Object Detection from Noisy Pseudo Label
di: Zhang, Jin, et al.
Pubblicazione: (2024)

Joint Asymmetric Loss for Learning with Noisy Labels
di: Wang, Jialiang, et al.
Pubblicazione: (2025)

Resilience of Vision Transformers for Domain Generalisation in the Presence of Out-of-Distribution Noisy Images
di: Riaz, Hamza, et al.
Pubblicazione: (2025)

Learning to Generate Diverse Pedestrian Movements from Web Videos with Noisy Labels
di: Liu, Zhizheng, et al.
Pubblicazione: (2024)

Multi-Label Out-of-Distribution Detection with Spectral Normalized Joint Energy
di: Mei, Yihan, et al.
Pubblicazione: (2024)

Rethinking Out-of-Distribution Detection on Imbalanced Data Distribution
di: Liu, Kai, et al.
Pubblicazione: (2024)

Out-of-Distribution Detection Based on Total Variation Estimation
di: Ma, Dabiao, et al.
Pubblicazione: (2026)

Learning with Noisy Labels: Interconnection of Two Expectation-Maximizations
di: Kim, Heewon, et al.
Pubblicazione: (2024)

Learning Transferable Negative Prompts for Out-of-Distribution Detection
di: Li, Tianqi, et al.
Pubblicazione: (2024)

Online Writer Retrieval with Chinese Handwritten Phrases: A Synergistic Temporal-Frequency Representation Learning Approach
di: Zhang, Peirong, et al.
Pubblicazione: (2024)

Towards Few-shot Out-of-Distribution Detection
di: Dong, Jiuqing, et al.
Pubblicazione: (2023)