SpecReason: Fast and Accurate Inference-Time Compute via Speculative Reasoning

Documenti analoghi

• SuffixDecoding: Extreme Speculative Decoding for Emerging AI Applications
  di: Oliaro, Gabriele, et al.
  Pubblicazione: (2024)
• Fail Fast, Win Big: Rethinking the Drafting Strategy in Speculative Decoding via Diffusion LLMs
  di: Pan, Rui, et al.
  Pubblicazione: (2025)
• Less Is More: Fast and Accurate Reasoning with Cross-Head Unified Sparse Attention
  di: Yang, Lijie, et al.
  Pubblicazione: (2025)
• Quantized Side Tuning: Fast and Memory-Efficient Tuning of Quantized Large Language Models
  di: Zhang, Zhengxin, et al.
  Pubblicazione: (2024)
• TidalDecode: Fast and Accurate LLM Decoding with Position Persistent Sparse Attention
  di: Yang, Lijie, et al.
  Pubblicazione: (2024)