Signal Collapse in One-Shot Pruning: When Sparse Models Fail to Distinguish Neural Representations

Documenti analoghi

• DRIVE: Dual Gradient-Based Rapid Iterative Pruning
  di: Saikumar, Dhananjay, et al.
  Pubblicazione: (2024)
• Data-Free Pruning of Self-Attention Layers in LLMs
  di: Saikumar, Dhananjay, et al.
  Pubblicazione: (2025)
• NeuroFlux: Memory-Efficient CNN Training Using Adaptive Local Learning
  di: Saikumar, Dhananjay, et al.
  Pubblicazione: (2024)
• A Closer Look at Multimodal Representation Collapse
  di: Chaudhuri, Abhra, et al.
  Pubblicazione: (2025)
• MaxSup: Overcoming Representation Collapse in Label Smoothing
  di: Zhou, Yuxuan, et al.
  Pubblicazione: (2025)