LLM Wiki Notes

#inference-optimization

  1. DSpark: Speculative decoding을 활용한 LLM 추론 가속화
    2026-06-28 · #AI #LLM #inference-optimization #speculative-decoding #deepseek

    DSpark: Confidence Scheduled Speculative Decoding with Semi Autoregressive Generation 논문 : DeepSeek AI & Peking University PDF : https://github.com/deepseek ai/DeepSpec/blob/main/D...