A Neural Passage Model for Ad-hoc Document Retrieval
This addresses the issue of irrelevant document parts in retrieval for users needing precise information, but it is incremental as it builds on existing passage-based paradigms.
The paper tackled the problem of ad-hoc document retrieval by proposing a neural passage model that uses passage-level information to improve performance, showing it significantly outperforms existing passage-based models on a TREC collection.
Traditional statistical retrieval models often treat each document as a whole. In many cases, however, a document is relevant to a query only because a small part of it contain the targeted information. In this work, we propose a neural passage model (NPM) that uses passage-level information to improve the performance of ad-hoc retrieval. Instead of using a single window to extract passages, our model automatically learns to weight passages with different granularities in the training process. We show that the passage-based document ranking paradigm from previous studies can be directly derived from our neural framework. Also, our experiments on a TREC collection showed that the NPM can significantly outperform the existing passage-based retrieval models.