The paper "Multi-hop Question Answering under Temporal Knowledge Editing" addresses the challenge of multi-hop question answering (MQA) in the context of knowledge editing (KE), particularly when dealing with questions containing explicit temporal contexts. The authors propose a novel framework called TEMPLE-MQA, which constructs a time-aware graph (TAG) to store edit knowledge in a structured manner. This framework enhances the ability to discern temporal contexts within questions and improves retrieval accuracy. TEMPLE-MQA employs an inference path, structural retrieval, and joint reasoning stages to effectively handle multi-hop questions with temporal scopes. Experimental results on benchmark datasets demonstrate that TEMPLE-MQA outperforms existing methods, and the authors also introduce a new dataset, TKEMQA, specifically designed for MQA with temporal scopes. The paper highlights the effectiveness of TEMPLE-MQA in maintaining historical knowledge while incorporating new edits, making it a significant advancement in the field of MQA under KE.The paper "Multi-hop Question Answering under Temporal Knowledge Editing" addresses the challenge of multi-hop question answering (MQA) in the context of knowledge editing (KE), particularly when dealing with questions containing explicit temporal contexts. The authors propose a novel framework called TEMPLE-MQA, which constructs a time-aware graph (TAG) to store edit knowledge in a structured manner. This framework enhances the ability to discern temporal contexts within questions and improves retrieval accuracy. TEMPLE-MQA employs an inference path, structural retrieval, and joint reasoning stages to effectively handle multi-hop questions with temporal scopes. Experimental results on benchmark datasets demonstrate that TEMPLE-MQA outperforms existing methods, and the authors also introduce a new dataset, TKEMQA, specifically designed for MQA with temporal scopes. The paper highlights the effectiveness of TEMPLE-MQA in maintaining historical knowledge while incorporating new edits, making it a significant advancement in the field of MQA under KE.