Replacing the reranker with the pruner completely was not feasable since it would have to run on the 150-200 chunks we retrieve before reranking.
Yes, the pruner operates on the top 15 chunks after reranking. reply
"relevance-to-the-query is exactly the signal that misses the buried caveat." You are exactly hitting the nail on the head. The underlaying problem of a pruner is to determine document relevance to a user query. And…
Replacing the reranker with the pruner completely was not feasable since it would have to run on the 150-200 chunks we retrieve before reranking.
Yes, the pruner operates on the top 15 chunks after reranking. reply
"relevance-to-the-query is exactly the signal that misses the buried caveat." You are exactly hitting the nail on the head. The underlaying problem of a pruner is to determine document relevance to a user query. And…