Thanks for your interesting and awesome work! I have carefully read your paper, but there is still two points that is unclear to me:
- In Section 2.2, what does the term "commercial LLM" specifically refer to? Is it different from the models mentioned in Section 3.2, such as Doubao-1.5-pro-32k, GPT-4o-mini, etc.?
- Additionally, are the prompts used in Section 2.2 consistent with those in Section A.5?
I would greatly appreciate your clarification on these issues. Looking forward to your reply.