In the process of providing PDF to text, everything else is quite good, it is in the prompt picture, did not extract the description of the picture description text.
Now the extraction results are like this:
![5_image_0.png](5_image_0.png)
In fact, it should be necessary to generate something like the following to meet expectations:
![The descrption text of the image should be here](5_image_0.png)
In the process of providing PDF to text, everything else is quite good, it is in the prompt picture, did not extract the description of the picture description text.
Now the extraction results are like this:
In fact, it should be necessary to generate something like the following to meet expectations: