DyLLM: Efficient Diffusion LLM Inference via Saliency-based Token Selection and Partial Attention

Documenti analoghi

• From Tokens to Layers: Redefining Stall-Free Scheduling for MoE Serving with Layered Prefill
  di: Lee, Gunjun, et al.
  Pubblicazione: (2025)
• An Inquiry into Datacenter TCO for LLM Inference with FP8
  di: Kim, Jiwoo, et al.
  Pubblicazione: (2025)
• Beyond Tokens: Semantic-Aware Speculative Decoding for Efficient Inference by Probing Internal States
  di: Dong, Ximing, et al.
  Pubblicazione: (2026)
• Dynamic Model Routing and Cascading for Efficient LLM Inference: A Survey
  di: Moslem, Yasmin, et al.
  Pubblicazione: (2026)
• Performance Characterization of Expert Router for Scalable LLM Inference
  di: Pichlmeier, Josef, et al.
  Pubblicazione: (2024)