A key component of BM25 contributing to its success is its sub-linear term frequency (TF) normalization formula. The scale and shape of this TF normalization component is controll...
We reveal that the Okapi BM25 retrieval function tends to overly penalize very long documents. To address this problem, we present a simple yet effective extension of BM25, namel...
In this paper, we reveal a common deficiency of the current retrieval models: the component of term frequency (TF) normalization by document length is not lower-bounded properly;...
In the robust track, we mainly tested our passage-based retrieval model with different passage sizes and weighting schemes. In our approach, we used two retrieval models, namely t...
Retrieval accuracy can be improved by considering which document type should be filtered out and which should be ranked higher in the result list. Hence, document type can be used...