Improving Document Representations by Generating Pseudo Query Embeddings for Dense Retrieval
This work addresses a specific bottleneck in dense retrieval for document search, offering an incremental improvement over existing methods.
The paper tackled the problem of information loss in dense retrieval models due to query-agnostic document encoding by generating pseudo query embeddings through iterative clustering and optimizing the matching function with a two-step score calculation, achieving state-of-the-art results on popular ranking and QA datasets.
Recently, the retrieval models based on dense representations have been gradually applied in the first stage of the document retrieval tasks, showing better performance than traditional sparse vector space models. To obtain high efficiency, the basic structure of these models is Bi-encoder in most cases. However, this simple structure may cause serious information loss during the encoding of documents since the queries are agnostic. To address this problem, we design a method to mimic the queries on each of the documents by an iterative clustering process and represent the documents by multiple pseudo queries (i.e., the cluster centroids). To boost the retrieval process using approximate nearest neighbor search library, we also optimize the matching function with a two-step score calculation procedure. Experimental results on several popular ranking and QA datasets show that our model can achieve state-of-the-art results.