Comment by jiwidi

I wonder, how different is this now from qwen trained rerankers like zeroentropy ?