Rhea: Role-aware Heuristic Episodic Attention for Conversational LLMs

Documenti analoghi

• Alloc-MoE: Budget-Aware Expert Activation Allocation for Efficient Mixture-of-Experts Inference
  di: Liu, Baihui, et al.
  Pubblicazione: (2026)
• Dovetail: A CPU/GPU Heterogeneous Speculative Decoding for LLM inference
  di: Zhang, Libo, et al.
  Pubblicazione: (2024)
• Sparrow: Text-Anchored Window Attention with Visual-Semantic Glimpsing for Speculative Decoding in Video LLMs
  di: Zhang, Libo, et al.
  Pubblicazione: (2026)
• GRASS: Gradient-based Adaptive Layer-wise Importance Sampling for Memory-efficient Large Language Model Fine-tuning
  di: Tian, Kaiyuan, et al.
  Pubblicazione: (2026)
• CTTA-T: Continual Test-Time Adaptation for Text Understanding via Teacher-Student with a Domain-aware and Generalized Teacher
  di: Liu, Tianlun, et al.
  Pubblicazione: (2025)