Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

为什么多模态关于规范格式的prompt不写在system中 #115

Open
munian08 opened this issue Feb 18, 2025 · 3 comments
Open

为什么多模态关于规范格式的prompt不写在system中 #115

munian08 opened this issue Feb 18, 2025 · 3 comments

Comments

@munian08
Copy link

def make_conversation(example):
        return {
            "prompt": [
                {"role": "system", "content": SYSTEM_PROMPT},
                {"role": "user", "content": example["problem"]},
            ],
        }

    QUESTION_TEMPLATE = "{Question}  Output the thinking process in <think> </think> and final answer (number) in <answer> </answer> tags."

    def make_conversation_image(example):
        return {
            "prompt": [
                {
                    "role": "user",
                    "content": [
                        {"type": "image"},
                        {"type": "text", "text": QUESTION_TEMPLATE.format(Question=example["problem"])},
                    ],
                },
            ],
        }
@Yukino256
Copy link

个人猜测,Qwen2vl的system prompt应该是You are Qwen XXXXXXXX XXXX 这个。如果加入system prompt也许对instruct的模型的训练造成不良影响,因为前面那一段的system prompt被替换了?或许base模型可以替换system prompt。

以上仅供猜测🥲🥲

@xingenju
Copy link

可以实验验证一下, 加入和不加入的效果对比

@Jia-py
Copy link

Jia-py commented Feb 19, 2025

这好像是因为qwen在huggingface上给的例子就没有放在system prompt中。

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants