InfLLM-V2-Long-Sparse-Base

Papers using or citing this model 1

• InfLLM-V2: Dense-Sparse Switchable Attention for Seamless Short-to-Long Adaptation 32 citations · 2025 · arXiv (Cornell University) · arXiv

  Long-sequence processing is a critical capability for modern large language models. However, the self-attention mechanism in the standard Transformer architecture faces severe computational and memory bottlenecks when processing long...