SLIM: Spuriousness Mitigation with Minimal Human Annotations

Documenti analoghi

• SUNY: A Visual Interpretation Framework for Convolutional Neural Networks from a Necessary and Sufficient Perspective
  di: Xuan, Xiwei, et al.
  Pubblicazione: (2023)
• ReME: A Data-Centric Framework for Training-Free Open-Vocabulary Segmentation
  di: Xuan, Xiwei, et al.
  Pubblicazione: (2025)
• CLImage: Human-Annotated Datasets for Complementary-Label Learning
  di: Wang, Hsiu-Hsuan, et al.
  Pubblicazione: (2023)
• Constructing Concept-based Models to Mitigate Spurious Correlations with Minimal Human Effort
  di: Kim, Jeeyung, et al.
  Pubblicazione: (2024)
• Focusing Image Generation to Mitigate Spurious Correlations
  di: Li, Xuewei, et al.
  Pubblicazione: (2024)