AdaKD: Dynamic Knowledge Distillation of ASR models using Adaptive Loss Weighting

Documenti analoghi

• FAST: Feature Aware Similarity Thresholding for Weak Unlearning in Black-Box Generative Models
  di: Panda, Subhodip, et al.
  Pubblicazione: (2023)
• Spectral Discovery of Continuous Symmetries via Generalized Fourier Transforms
  di: Karjol, Pavan, et al.
  Pubblicazione: (2026)
• AdaGMLP: AdaBoosting GNN-to-MLP Knowledge Distillation
  di: Lu, Weigang, et al.
  Pubblicazione: (2024)
• S^2-KD: Semantic-Spectral Knowledge Distillation Spatiotemporal Forecasting
  di: Wang, Wenshuo, et al.
  Pubblicazione: (2025)
• Leveraging Data Symmetries to Select an Optimal Subset of Training Data under Label Noise
  di: Shubham, Kumar, et al.
  Pubblicazione: (2026)