A Neural Passage Model for Ad-hoc Document Retrieval

TL;DR

This paper introduces a neural passage model that enhances ad-hoc document retrieval by automatically weighting passages of varying granularities, leading to significant performance improvements over existing models.

Contribution

The paper presents a novel neural passage model that learns to weight passages of different sizes, directly deriving from and improving upon previous passage-based retrieval frameworks.

Findings

NPM significantly outperforms existing passage-based retrieval models.

The model effectively learns to weight passages of different granularities.

Experimental results on TREC collection validate the approach.

Abstract

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.