ReasonBERT: Pre-trained to Reason with Distant Supervision

TL;DR

ReasonBert is a pre-training approach that enhances language models with long-range reasoning abilities by automatically connecting multiple texts and tables through distant supervision, improving performance on complex question answering tasks.

Contribution

It introduces a novel distant supervision method for pre-training that enables models to perform long-range and multi-modal reasoning, surpassing existing local-context-based methods.

Findings

Significant improvements on multi-hop question answering datasets.

Enhanced sample efficiency in few-shot learning scenarios.

Effective reasoning over hybrid text-table contexts.

Abstract

We present ReasonBert, a pre-training method that augments language models with the ability to reason over long-range relations and multiple, possibly hybrid contexts. Unlike existing pre-training methods that only harvest learning signals from local contexts of naturally occurring texts, we propose a generalized notion of distant supervision to automatically connect multiple pieces of text and tables to create pre-training examples that require long-range reasoning. Different types of reasoning are simulated, including intersecting multiple pieces of evidence, bridging from one piece of evidence to another, and detecting unanswerable cases. We conduct a comprehensive evaluation on a variety of extractive question answering datasets ranging from single-hop to multi-hop and from text-only to table-only to hybrid that require various reasoning capabilities and show that ReasonBert achieves remarkable improvement over an array of strong baselines. Few-shot experiments further demonstrate that our pre-training method substantially improves sample efficiency.