M2R2: Mixture of Multi-Rate Residuals for Efficient Transformer Inference

Documenti analoghi

• Speculative Streaming: Fast LLM Inference without Auxiliary Models
  di: Bhendawade, Nikhil, et al.
  Pubblicazione: (2024)
• FS-DFM: Fast and Accurate Long Text Generation with Few-Step Diffusion Language Models
  di: Monsefi, Amin Karimi, et al.
  Pubblicazione: (2025)
• Trajectory as the Teacher: Few-Step Discrete Flow Matching via Energy-Navigated Distillation
  di: Monsefi, Amin Karimi, et al.
  Pubblicazione: (2026)
• DACA-GRPO: Denoising-Aware Credit Assignment for Reinforcement Learning in Diffusion Language Models
  di: Monsefi, Amin Karimi, et al.
  Pubblicazione: (2026)
• LazyLLM: Dynamic Token Pruning for Efficient Long Context LLM Inference
  di: Fu, Qichen, et al.
  Pubblicazione: (2024)