Two are better than one: Context window extension with multi-grained self-injection

Documenti analoghi

• Stacked from One: Multi-Scale Self-Injection for Context Window Extension
  di: Han, Wei, et al.
  Pubblicazione: (2026)
• Towards Robust Instruction Tuning on Multimodal Large Language Models
  di: Han, Wei, et al.
  Pubblicazione: (2024)
• Reasoning Paths Optimization: Learning to Reason and Explore From Diverse Paths
  di: Chia, Yew Ken, et al.
  Pubblicazione: (2024)
• MOOSE-Chem: Large Language Models for Rediscovering Unseen Chemistry Scientific Hypotheses
  di: Yang, Zonglin, et al.
  Pubblicazione: (2024)
• Language models are better than humans at next-token prediction
  di: Shlegeris, Buck, et al.
  Pubblicazione: (2022)