Unveiling Language Routing Isolation in Multilingual MoE Models for Interpretable Subnetwork Adaptation

Documenti analoghi

• Dynamic Expert Specialization: Towards Catastrophic Forgetting-Free Multi-Domain MoE Adaptation
  di: Li, Junzhuo, et al.
  Pubblicazione: (2025)
• Reefknot: A Comprehensive Benchmark for Relation Hallucination Evaluation, Analysis and Mitigation in Multimodal Large Language Models
  di: Zheng, Kening, et al.
  Pubblicazione: (2024)
• Deconstructing Pre-training: Knowledge Attribution Analysis in MoE and Dense Models
  di: Wang, Bo, et al.
  Pubblicazione: (2026)
• GM-PRM: A Generative Multimodal Process Reward Model for Multimodal Mathematical Reasoning
  di: Zhang, Jianghangfan, et al.
  Pubblicazione: (2025)
• Pierce the Mists, Greet the Sky: Decipher Knowledge Overshadowing via Knowledge Circuit Analysis
  di: Huang, Haoming, et al.
  Pubblicazione: (2025)