ReasonBERT: Pre-trained to Reason with Distant Supervision
This work addresses the challenge of improving reasoning capabilities in language models for tasks like multi-hop question answering, representing an incremental advance over existing pre-training methods.
The paper tackles the problem of enabling language models to perform long-range reasoning across multiple contexts by introducing ReasonBERT, a pre-training method that uses distant supervision to create examples requiring complex reasoning. The result is a model that achieves significant improvements on various extractive question answering datasets and shows enhanced sample efficiency in few-shot experiments.
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.