AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Impact of Noisy Supervision in Foundation Model Learning

Hao Chen, Zihan Wang, Ran Tao, Hongxin Wei, Xing Xie, Masashi Sugiyama, Bhiksha Raj, Jindong Wang

TPAMI 2025 · Label-efficient and robust learning › Recognition and long-tail · approach: Data augmentation and generalization bias · 14 citations (OpenAlex)

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
add NMTune regularization to linear probingViT-L (Laion-2B CLIP) · 14 ID vision datasetsaccuracy: 86.85 -> 88.57Improved
add NMTune regularization to linear probingConvNext-L (Laion-2B CLIP) · 4 OOD vision datasetsaccuracy: 66.86 -> 70.30Improved
add regularization terms to MLP tuning (ablation)ViT-L · 4 OOD vision datasetsaccuracy: 69.50 -> 70.47Improved
add an extra MLP head alone (no regularization) to LP (ablation)ConvNext-L · 14 ID vision datasetsaccuracy: 88.86 -> 88.53Hurt
add NMTune to API embedding modeltext-ada-002 · GLUE-X (OOD)score: 44.06 -> 53.48Improved
add NMTune to MLP tuning on GLUE (ablation)BERT-L · GLUE-X (OOD)score: 50.62 -> 51.63Improved
raise pre-training label noise from 0% to 5%ResNet-50 / ViT-B-16 (IN-1K FS, CLIP) · downstream ID classification (LP)accuracyImproved
raise pre-training label noise from 0% to 5%ResNet-50 / ViT-B-16 · downstream OOD classification (LP, LoRA)accuracyHurt
raise pre-training noise from 0% to 30%ViT-B-16 IN-1K FS · IN-1K validationaccuracy: 78.70 -> 68.12Hurt
raise pre-training noise from 0% to 5%Faster R-CNN with ResNet-50 · COCO 2017 detectionAP box: 38.5 -> 38.6No clear effect
raise pre-training noise from 0% to 30%Mask R-CNN with ResNet-50 · COCO 2017 segmentationAP mask: 31.3 -> 30.3Hurt
tune more parameters (LP to LoRA to full fine-tuning)noisy pre-trained ViT-B-16 · ID and OOD classificationaccuracyImproved

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