BAAI

/

bge-large-en-v1.5

like 449

Follow

Beijing Academy of Artificial Intelligence 593

Hi, the instruction for the query is: Represent this sentence for searching relevant passages:

Hello, just wondering. We dont need instruction query though correct? I am using sentencetransformers, and this is important. I have Long query, and my passages are short, Imagine a customer complaint, and then a simple short comment that is picked that best matches the Long user query. Here is an example

User query 500 tokens long

10 short comments all 50-100 words

The model encodes the user query and the comments, and then picks the best short comment that matches the long query. My guess is in this case we do not need to use instructions?

Thanks

Hello, just wondering. We dont need instruction query though correct? I am using sentencetransformers, and this is important. I have Long query, and my passages are short, Imagine a customer complaint, and then a simple short comment that is picked that best matches the Long user query. Here is an example

User query 500 tokens long

10 short comments all 50-100 words

The model encodes the user query and the comments, and then picks the best short comment that matches the long query. My guess is in this case we do not need to use instructions?

Thanks

Hi, I think there is no need to use instructions. You also can try both settings and select which can achieve better performance.