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

[Bug] 使用GPT模型,第二张图片就无法正常识别了 #4308

Open
Ray1735 opened this issue Oct 9, 2024 · 4 comments
Open

[Bug] 使用GPT模型,第二张图片就无法正常识别了 #4308

Ray1735 opened this issue Oct 9, 2024 · 4 comments
Labels
🐛 Bug Something isn't working | 缺陷

Comments

@Ray1735
Copy link

Ray1735 commented Oct 9, 2024

📦 Environment

Official Preview

📌 Version

GPT-4o 0513/GPT-4o 0806/ChatGPT-4o

💻 Operating System

Windows

🌐 Browser

Chrome, Firefox

🐛 Bug Description

每个新话题,当我要求描述第一张发送的图片,都可以准确回答图片的内容。当我要求描述第二张发送的图片,回答还是会继续描述第一张图片,或者直接回答“抱歉,我无法查看或描述该图片的内容。”,偶会会描述不存在的图片。需要多次提问后,才会识别第二张图片。
我试过用谷歌浏览器和火狐浏览器都是同样的问题,不同的WIN10电脑设备也有这个问题。切换不同的GPT版本,例如GPT-4o 0513/GPT-4o 0806/ChatGPT-4o,还是会出现这个问题。
65f85ae93628d877ad89b4c3948f913

📷 Recurrence Steps

No response

🚦 Expected Behavior

No response

📝 Additional Information

No response

@Ray1735 Ray1735 added the 🐛 Bug Something isn't working | 缺陷 label Oct 9, 2024
@lobehubbot
Copy link
Member

👀 @Ray1735

Thank you for raising an issue. We will investigate into the matter and get back to you as soon as possible.
Please make sure you have given us as much context as possible.
非常感谢您提交 issue。我们会尽快调查此事,并尽快回复您。 请确保您已经提供了尽可能多的背景信息。

Copy link

dosubot bot commented Oct 9, 2024

Hey @Ray1735, I'm here to help you with any bugs, questions, or contributions you have! Let's squash those bugs together! 🤖

The issue you're experiencing with the GPT model not recognizing the second image correctly might be related to how the system differentiates and processes images in a conversation. The system uses an imageList in the message content to manage images, where each image has an id, url, and alt text. This setup should allow the system to handle multiple images [1].

However, if the model continues to describe the first image or fails to recognize the second one, it could be a bug in the image context management. There might be a need to review the logic that updates the context when a new image is processed, as this could be causing the issue.

To address this, you might want to check if the image IDs are being correctly updated and if the context-switching logic is functioning as expected. Additionally, ensure that the error handling mechanisms are logging any issues that occur during image processing [2].

To continue talking to Dosu, mention @dosu.

@Ray1735
Copy link
Author

Ray1735 commented Oct 10, 2024

我不是程序员,只是一个普通用户,没能力分析这些问题呢。但是我用F12检查,发现输入图片时并不会报错。

@lobehubbot
Copy link
Member

Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑‍🤝‍🧑👫🧑🏿‍🤝‍🧑🏻👩🏾‍🤝‍👨🏿👬🏿


I'm not a programmer, just an ordinary user, and I don't have the ability to analyze these problems. But I checked with F12 and found that no error was reported when inputting pictures.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
🐛 Bug Something isn't working | 缺陷
Projects
Status: Roadmap - Chat 1.x
Development

No branches or pull requests

2 participants