Decoding Knowledge Attribution in Mixture-of-Experts: A Framework of Basic-Refinement Collaboration and Efficiency Analysis

Documenti analoghi

• Dynamic Expert Specialization: Towards Catastrophic Forgetting-Free Multi-Domain MoE Adaptation
  di: Li, Junzhuo, et al.
  Pubblicazione: (2025)
• Optimal Expert-Attention Allocation in Mixture-of-Experts: A Scalable Law for Dynamic Model Design
  di: Li, Junzhuo, et al.
  Pubblicazione: (2026)
• Deconstructing Pre-training: Knowledge Attribution Analysis in MoE and Dense Models
  di: Wang, Bo, et al.
  Pubblicazione: (2026)
• Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression
  di: Liu, Xiang, et al.
  Pubblicazione: (2025)
• Not All Experts are Equal: Efficient Expert Pruning and Skipping for Mixture-of-Experts Large Language Models
  di: Lu, Xudong, et al.
  Pubblicazione: (2024)