Autoregressive Reasoning over Chains of Facts with Transformers

Abstract

This paper proposes an iterative inference algorithm for multi-hop explanation regeneration, that retrieves relevant factual evidence in the form of text snippets, given a natural language question and its answer. Combining multiple sources of evidence or facts for multi-hop reasoning becomes increasingly hard when the number of sources needed to make an inference grows. Our algorithm copes with this by decomposing the selection of facts from a corpus autoregressively, conditioning the next iteration on previously selected facts. This allows us to use a pairwise learning-to-rank loss. We validate our method on datasets of the TextGraphs 2019 and 2020 Shared Tasks for explanation regeneration. Existing work on this task either evaluates facts in isolation or artificially limits the possible chains of facts, thus limiting multi-hop inference. We demonstrate that our algorithm, when used with a pre-trained transformer model, outperforms the previous state-of-the-art in terms of precision, training time and inference efficiency.

arXiv commentPublished at International Conference on Computational Linguistics 2020 (ICCL) (COLING)

References in corpus (9)

• Scikit-learn: Machine Learning in Python
• Distributed Representations of Words and Phrases and their Compositionality
• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• HuggingFace's Transformers: State-of-the-art Natural Language Processing
• Scheduled Sampling for Sequence Prediction with Recurrent Neural Networks
• Passage Re-ranking with BERT
• Multi-step Retriever-Reader Interaction for Scalable Open-domain Question Answering
• From 'F' to 'A' on the N.Y. Regents Science Exams: An Overview of the Aristo Project
• Red Dragon AI at TextGraphs 2019 Shared Task: Language Model Assisted Explanation Generation