Thank you for your interesting and meaningful work. But I still have some questions about the experimental part of scientific report generation. I noticed that the article mentioned three types of RAG: Standard RAG, RAG w/ Query Planning and Iterative RAG. Do you use Standard RAG when evaluating Qwen2.5-72B and DeepSeek-R1 on scientific report generation task? Could you please provide the generation results of these two baselines? Thank you very much