One Size Does Not Fit All: A Distribution-Aware Sparsification for More Precise Model Merging

Documenti analoghi

• NiuTrans.LMT: Toward Inclusive and Scalable Multilingual Machine Translation with LLMs
  di: Luo, Yingfeng, et al.
  Pubblicazione: (2025)
• RouteLMT: Learned Sample Routing for Hybrid LLM Translation Deployment
  di: Luo, Yingfeng, et al.
  Pubblicazione: (2026)
• Efficient Prompting Methods for Large Language Models: A Survey
  di: Chang, Kaiyan, et al.
  Pubblicazione: (2024)
• APR: Penalizing Structural Redundancy in Large Reasoning Models via Anchor-based Process Rewards
  di: Chang, Kaiyan, et al.
  Pubblicazione: (2026)
• Early Exit Is a Natural Capability in Transformer-based Models: An Empirical Study on Early Exit without Joint Optimization
  di: Shan, Weiqiao, et al.
  Pubblicazione: (2024)