Wang Baoling

icon indicating an email [email protected]

Results 4 comments of Wang Baoling

[Feature Request]: Markdown and JSON Support

+1

[Bug]: The size of PDF chunks is not limited, which can lead to extremely large chunks that exceed the limitations of the LLM in extreme cases.

A PDF file is parsed and splited to chunks, each chunk's size should be limited.

[Bug]: For text PDF files, OCR reduces the recognition rate. It is recommended to add an option to enable OCR in the 【Chunk Method】if can not do it.

> And there's lack of method to judge it good enough or not. In this case, you can provide an option and let users judge it

[Bug]: For text PDF files, OCR reduces the recognition rate. It is recommended to add an option to enable OCR in the 【Chunk Method】if can not do it.

> Accturaly, it is not the result from OCR. > If text can be read from PDF, it will not bother using OCR. _**the param chars is text from PDF...