Rethinking Calibration for Early-Exit Neural Networks

Documenti analoghi

• How to Train Your Multi-Exit Model? Analyzing the Impact of Training Strategies
  di: Kubaty, Piotr, et al.
  Pubblicazione: (2024)
• Exploiting Activation Sparsity with Dense to Dynamic-k Mixture-of-Experts Conversion
  di: Szatkowski, Filip, et al.
  Pubblicazione: (2023)
• Early-Exit Neural Networks with Nested Prediction Sets
  di: Jazbec, Metod, et al.
  Pubblicazione: (2023)
• Efficient LLM Moderation with Multi-Layer Latent Prototypes
  di: Chrabąszcz, Maciej, et al.
  Pubblicazione: (2025)
• Universal Properties of Activation Sparsity in Modern Large Language Models
  di: Szatkowski, Filip, et al.
  Pubblicazione: (2025)