← Back to context Comment by bardiapour 1 year ago i think not, better results >>> better latency + cost 1 comment bardiapour Reply robrenaud 1 year ago Maybe a combined approach beats either? Let some non-LLM reranker quickly spit out two results, and fill in the rest with the LLM.
robrenaud 1 year ago Maybe a combined approach beats either? Let some non-LLM reranker quickly spit out two results, and fill in the rest with the LLM.
Maybe a combined approach beats either? Let some non-LLM reranker quickly spit out two results, and fill in the rest with the LLM.