AI Paper Atlas

Research map / Label-efficient and robust learning / Noisy labels

Mitigating Label Noise using Prompt-Based Hyperbolic Meta-Learning in Open-Set Domain Generalization

Kunyu Peng, Di Wen, Muhammad Saquib Sarfraz, Yufan Chen, Junwei Zheng, David Schneider, Kailun Yang, Jiamin Wu, Alina Roitberg, Rainer Stiefelhagen

IJCV 2026 · Label-efficient and robust learning › Noisy labels · approach: Noise-aware loss and pairwise methods · 4 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
use HyProMeta instead of best OSDG baseline MEDIC-bclsMEDIC-bcls · PACS ResNet18, symmetric noise 20%OSCR: 42.76 -> 48.20Improved
use HyProMeta instead of MEDIC-bcls under heavy noiseMEDIC-bcls · PACS ResNet18, symmetric noise 80%OSCR: 11.17 -> 25.09Improved
use HyProMeta instead of MEDIC-bcls under asymmetric noiseMEDIC-bcls · PACS ResNet18, asymmetric noise 50%OSCR: 23.93 -> 37.80Improved
use HyProMeta instead of MLDGMLDG · PACS ViT-Base, symmetric noise 50%OSCR: 49.64 -> 52.91Improved
use HyProMeta instead of EBiL-HaDS-bclsEBiL-HaDS-bcls · PACS ViT-Base, symmetric noise 80%OSCR: 16.90 -> 25.26Improved
use HyProMeta instead of MLDGMLDG · PACS ViT-Base, asymmetric noise 50%OSCR: 40.70 -> 43.44Improved
raise symmetric noise ratio from 20% to 80%HyProMeta · PACS ResNet18OSCR: 48.20 -> 25.09Hurt
switch backbone from ResNet18 to ViT-BaseHyProMeta · PACS symmetric noise 20%OSCR: 48.20 -> 54.97Improved

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