Functional Interpolation for Relative Positions Improves Long Context Transformers

Documenti analoghi

• Scalable In-context Ranking with Generative Models
  di: Gupta, Nilesh, et al.
  Pubblicazione: (2025)
• Position Coupling: Improving Length Generalization of Arithmetic Transformers Using Task Structure
  di: Cho, Hanseul, et al.
  Pubblicazione: (2024)
• HiRE: High Recall Approximate Top-$k$ Estimation for Efficient LLM Inference
  di: L, Yashas Samaga B, et al.
  Pubblicazione: (2024)
• On student-teacher deviations in distillation: does it pay to disobey?
  di: Nagarajan, Vaishnavh, et al.
  Pubblicazione: (2023)
• Mimetic Initialization Helps State Space Models Learn to Recall
  di: Trockman, Asher, et al.
  Pubblicazione: (2024)