Megalodon: Efficient LLM Pretraining and Inference with Unlimited Context Length

Documenti analoghi

• Transfusion: Predict the Next Token and Diffuse Images with One Multi-Modal Model
  di: Zhou, Chunting, et al.
  Pubblicazione: (2024)
• LMFusion: Adapting Pretrained Language Models for Multimodal Generation
  di: Shi, Weijia, et al.
  Pubblicazione: (2024)
• Self-Alignment with Instruction Backtranslation
  di: Li, Xian, et al.
  Pubblicazione: (2023)
• Gecko: An Efficient Neural Architecture Inherently Processing Sequences with Arbitrary Lengths
  di: Ma, Xuezhe, et al.
  Pubblicazione: (2026)
• CAT: Content-Adaptive Image Tokenization
  di: Shen, Junhong, et al.
  Pubblicazione: (2025)