CAS-Spec: Cascade Adaptive Self-Speculative Decoding for On-the-Fly Lossless Inference Acceleration of LLMs

Documenti analoghi

• Pipeline Parallelism is All You Need for Optimized Early-Exit Based Self-Speculative Decoding
  di: Li, Ruanjun, et al.
  Pubblicazione: (2025)
• LongSpec: Long-Context Lossless Speculative Decoding with Efficient Drafting and Verification
  di: Yang, Penghui, et al.
  Pubblicazione: (2025)
• SWIFT: On-the-Fly Self-Speculative Decoding for LLM Inference Acceleration
  di: Xia, Heming, et al.
  Pubblicazione: (2024)
• Accelerating LLM Inference with Lossless Speculative Decoding Algorithms for Heterogeneous Vocabularies
  di: Timor, Nadav, et al.
  Pubblicazione: (2025)
• BanditSpec: Adaptive Speculative Decoding via Bandit Algorithms
  di: Hou, Yunlong, et al.
  Pubblicazione: (2025)