Towards Interpretable and Efficient Attention: Compressing All by Contracting a Few

Documenti analoghi

• Mixture-of-Top-k Attention: Efficient Attention via Scalable Fast Weights
  di: Wen, Qishuai, et al.
  Pubblicazione: (2026)
• Rethinking Decoders for Transformer-based Semantic Segmentation: A Compression Perspective
  di: Wen, Qishuai, et al.
  Pubblicazione: (2024)
• Exploring a Principled Framework for Deep Subspace Clustering
  di: Meng, Xianghan, et al.
  Pubblicazione: (2025)
• EDiT: Efficient Diffusion Transformers with Linear Compressed Attention
  di: Becker, Philipp, et al.
  Pubblicazione: (2025)
• COMCAT: Towards Efficient Compression and Customization of Attention-Based Vision Models
  di: Xiao, Jinqi, et al.
  Pubblicazione: (2023)