Equifinality in Mixture of Experts: Routing Topology Does Not Determine Language Modeling Quality

Documenti analoghi

• Geometric Routing Enables Causal Expert Control in Mixture of Experts
  di: Ternovtsii, Ivan, et al.
  Pubblicazione: (2026)
• Cosine-Similarity Routing with Semantic Anchors for Interpretable Mixture-of-Experts Language Models
  di: Ternovtsii, Ivan, et al.
  Pubblicazione: (2025)
• Mixture of Layers with Hybrid Attention
  di: Ternovtsii, Ivan, et al.
  Pubblicazione: (2026)
• AdaMoE: Token-Adaptive Routing with Null Experts for Mixture-of-Experts Language Models
  di: Zeng, Zihao, et al.
  Pubblicazione: (2024)
• Routing-Free Mixture-of-Experts
  di: Liu, Yilun, et al.
  Pubblicazione: (2026)