L3: DIMM-PIM Integrated Architecture and Coordination for Scalable Long-Context LLM Inference

Re-assign community