Multi-Lingual Question Generation with Language Agnostic Language Model

Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021, pages 1–12 August 1–6, 2021. ©2021 Association for Computational Linguistics 1 Explainable Inference Over Grounding-Abstract Chains for Science Questions Mokanarangan Thayaparan†, Marco Valentino†, Andr´e Freitas†‡ Department of Computer Science, University of Manchester, United Kingdom† Idiap Research Institute, Switzerland‡ {firstname.lastname}@manchester.ac.uk Abstract We propose an explainable inference approach for science questions by reasoning on ground- ing and abstract inference chains. This paper frames question answering as a natural lan- guage abductive reasoning problem, construct- ing plausible explanations for each candidate answer and then selecting the candidate with the best explanation as the final answer. Our method, ExplanationLP, elicits explanations by constructing a weighted graph of relevant facts for each candidate answer and employs a linear programming formalism designed to select the optimal subgraph of explanatory facts. The graphs’ weighting function is com- posed of a set of parameters targeting rele- vance, cohesion and diversity, which we fine- tune for answer selection via Bayesian Opti- misation. We carry out our experiments on the WorldTree and ARC-Challenge datasets to em- pirically demonstrate the following contribu- tions: (1) ExplanationLP obtains strong perfor- mance when compared to transformer-based and multi-hop approaches despite having a sig- nificantly lower number of parameters; (2) We show that our model is able to generate plausi- ble explanations for answer prediction; (3) Our model demonstrates better robustness towards semantic drift when compared to transformer- based and multi-hop approaches. 1 Introduction Answering science questions remain a fundamen- tal challenge in Natural Language Processing and AI as it requires complex forms of inference, in- cluding causal, model-based and example-based reasoning (Jansen, 2018; Clark et al., 2018; Jansen et al., 2016; Clark et al., 2013). Current state-of-the- art (SOTA) approaches for answering questions in the science domain are dominated by transformer- based models (Devlin et al., 2019; Sun et al., 2019). Despite remarkable performance on answer pre- diction, these approaches are black-box by nature, lacking the capability of providing explanations for their predictions (Thayaparan et al., 2020; Miller, 2019; Biran and Cotton, 2017; Jansen et al., 2016). Explainable Science Question Answering (XSQA) is often framed as a natural language abductive reasoning problem (Khashabi et al., 2018; Jansen et al., 2017). Abductive reasoning represents a distinct inference process, known as inference to the best explanation (Peirce, 1960; Lipton, 2017), which starts from a set of complete or incomplete observations to find the hypothesis, from a set of plausible alternatives, that best explains the observations. Several approaches (Khashabi et al., 2018; Jansen et al., 2017; Khot et al., 2017a; Khashabi et al., 2016) employ this form of reasoning for multiple-choice science questions to build a set of plausible explanations for each candidate answer and select the one with the best explanation as the final answer. XSQA solvers typically treat explanation gener- ation as a multi-hop graph traversal problem. Here, the solver attempts to compose multiple facts that connect the question to a candidate answer. These multi-hop approaches have shown diminishing re- turns with an increasing number of …

Content truncated to 500 words