NORA applies task-aware weighting and NPK filtering to handle label noise in multi-attribute tagging of financial numerical entities, outperforming baselines on a new 6.6M-instance benchmark.
Learning From Noisy Labels With Deep Neural Networks: A Survey
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 5verdicts
UNVERDICTED 5roles
background 1polarities
background 1representative citing papers
Calibrates the noisy-label crossover for BiomedCLIP weak supervision on PCAM, ISIC and NIH-CXR, reports architecture-invariant locations, and derives a decision rule comparing gold-only AUC to VLM accuracy.
BrainSimSiam applies positive-only Siamese self-supervised learning to fMRI data to produce representations that generalize across downstream tasks and outperform supervised baselines.
Proposes the EL-MIATTs framework for ML predictive modeling by assuming the true target does not exist and defining democratic supervision via multiple inaccurate true targets.
The EL-MIATTs framework offers LAF-grounded evaluation and UTTL-grounded learning strategies to handle multiple inaccurate true targets in machine learning under uncertain supervision.
citing papers explorer
-
From Theory to Decision Rule: Calibrating the Noisy-Label Crossover for Vision-Language Model Weak Supervision Across Three Medical-Imaging Benchmarks
Calibrates the noisy-label crossover for BiomedCLIP weak supervision on PCAM, ISIC and NIH-CXR, reports architecture-invariant locations, and derives a decision rule comparing gold-only AUC to VLM accuracy.