Data Curation Matters: Model Collapse and Spurious Shift Performance Prediction from Training on Uncurated Text Embeddings

Documenti analoghi

• LACON: Training Text-to-Image Model from Uncurated Data
  di: Liang, Zhiyang, et al.
  Pubblicazione: (2026)
• Your Attention Matters: to Improve Model Robustness to Noise and Spurious Correlations
  di: Tamayo-Rousseau, Camilo, et al.
  Pubblicazione: (2025)
• Training on Plausible Counterfactuals Removes Spurious Correlations
  di: Sadiku, Shpresim, et al.
  Pubblicazione: (2025)
• Bridging Explainability and Embeddings: BEE Aware of Spuriousness
  di: Păduraru, Cristian Daniel, et al.
  Pubblicazione: (2024)
• Complexity Matters: Dynamics of Feature Learning in the Presence of Spurious Correlations
  di: Qiu, GuanWen, et al.
  Pubblicazione: (2024)