<Evaluation of prompts>

[TongLi97]

是否已有关于该错误的issue或讨论？ | Is there an existing issue / discussion for this?

• 我已经搜索过已有的issues和讨论 | I have searched the existing issues / discussions

该问题是否在FAQ中有解答？ | Is there an existing answer for this in FAQ?

• 我已经搜索过FAQ | I have searched FAQ

当前行为 | Current Behavior

I want to test the effect of different prompts on Qwen-VL, but not having a rule makes it difficult for me to formulate prompts.
Is there any documentation on the impact of different prompt instructions on Qwen-VL output?
I hope to use Qwen in my research, so it is important to me that there are relevant rules and assessments.

期望行为 | Expected Behavior

No response

复现方法 | Steps To Reproduce

No response

运行环境 | Environment

No response

备注 | Anything else?

No response

[ShuaiBai623]

Thank you very much for your attention. Different prompts have a great impact on the results. We provide some simple examples in TUTORIAL.md. However, we haven’t done more in-depth exploration of prompts yet, and we're looking forward to your results.

[TongLi97]

Thank you very much for your reply.&nbsp; Can your organization provide the instruction data set used in the training process (mentioned in the Qwen or Qwen-VL technical report)? Or tell me the public instruction data set used so that I can download and understand it. Thank you so much.

…

------------------&nbsp;原始邮件&nbsp;------------------ 发件人: "QwenLM/Qwen-VL" ***@***.***&gt;; 发送时间:&nbsp;2023年10月25日(星期三) 晚上8:19 ***@***.***&gt;; ***@***.******@***.***&gt;; 主题:&nbsp;Re: [QwenLM/Qwen-VL] <Evaluation of prompts&gt; (Issue #135) 非常感谢您的关注。不同的提示对结果影响很大。我们在 TUTORIAL.md 中提供了一些简单的示例。不过，我们还没有对提示进行更深入的探索，我们期待您的结果。 — 直接回复此电子邮件、在 GitHub 上查看或取消订阅。 您收到此消息是因为您创作了该主题。消息 ID：<QwenLM/Qwen-VL/issues/135/1779149430 @ github 。com&gt;