Larger-Context Tagging: When and Why Does It Work?

TL;DR

This paper investigates the conditions under which larger-context training improves sentence-level tagging systems, analyzing different aggregation methods and providing interpretability to guide future use of contextual information.

Contribution

It offers a comprehensive study on when and why larger-context training works, including a comparative analysis of aggregation methods and an interpretability framework.

Findings

Larger-context training benefits depend on specific task attributes.

Different aggregation methods vary in effectiveness across datasets.

The interpretability method clarifies when larger-context improves performance.

Abstract

The development of neural networks and pretraining techniques has spawned many sentence-level tagging systems that achieved superior performance on typical benchmarks. However, a relatively less discussed topic is what if more context information is introduced into current top-scoring tagging systems. Although several existing works have attempted to shift tagging systems from sentence-level to document-level, there is still no consensus conclusion about when and why it works, which limits the applicability of the larger-context approach in tagging tasks. In this paper, instead of pursuing a state-of-the-art tagging system by architectural exploration, we focus on investigating when and why the larger-context training, as a general strategy, can work. To this end, we conduct a thorough comparative study on four proposed aggregators for context information collecting and present an attribute-aided evaluation method to interpret the improvement brought by larger-context training. Experimentally, we set up a testbed based on four tagging tasks and thirteen datasets. Hopefully, our preliminary observations can deepen the understanding of larger-context training and enlighten more follow-up works on the use of contextual information.