RIS-Kernel: A Model-Agnostic Architecture for Long-Context LLM Inference via Sparse Attention Paper • 2607.21927 • Published Jul 24 • 1 • 2
RIS-Kernel: A Model-Agnostic Architecture for Long-Context LLM Inference via Sparse Attention Paper • 2607.21927 • Published Jul 24 • 1 • 2