WorldTree V2: A Corpus of Science-Domain Structured Explanations and Inference Patterns supporting Multi-Hop Inference

By Zhengnan Xie, Sebastian Thiem, Jaycie Martin, Elizabeth Wainwright, Steven Marmorstein, Peter Alexander Jansen. 2020

90 citations recorded by the source corpus.

Subfields: Natural Language Processing, Evaluation & Benchmarks

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Natural Language Processing

In Evaluation & Benchmarks

Citations in the corpus

Flexible Generation of Natural Language Deductions (2021, background): “More recent datasets emphasize the ability to actually exhibit correct reasoning chains and form explanations (Clark et al., 2020; Xie et al., 2020; Dalvi et al., 2021).”
Faithful Reasoning Using Large Language Models (2022, background): “However, in future work we hope to incorporate retrieval to populate the context, and there is already interesting research in this direction (Dalvi et al., 2021; Ribeiro et al., 2022; Xie et al., 2020).”
Explaining Answers with Entailment Trees (2021, methodology): “E NTAILMENT B ANK uses multiple-choice questions (and the correct answer option) from the ARC dataset of grade-school science questions (Clark et al., 2018), and a corpus of science-and general knowledge derived from WorldTree V2 (Xie et al., 2020; Jansen et al., 2018).”
Explanations for CommonsenseQA: New Dataset and Models (2021, background): “WorldTree (Jansen et al., 2018) and WorldTree V2 (Xie et al., 2020) are corpora of elementary multiple-choice science questions with gold explanations for correct answer choice.”
Think you have Solved Direct-Answer Question Answering? Try ARC-DA, the Direct-Answer AI2 Reasoning Challenge (2021, background): “, (Musa et al., 2019; Boratko et al., 2018; Ni et al., 2019; Xie et al., 2020), and as of January 2021 has 35 entries on its leaderboard2.”
Case-Based Abductive Natural Language Inference (2020, background): “…2020; Yang et al., 2018), increasing attention is being dedicated to the abstractive setting, where the models are required to compose long chains of facts expressing abstract commonsense and scientific knowledge (Jhamtani and Clark, 2020; Xie et al., 2020; Khot et al., 2019; Jansen et al., 2018).”
Entailment Tree Explanations via Iterative Retrieval-Generation Reasoner (2022, background): “Conditioning the retrieval of a passage on previously retrieved passages has been explored in the context of multi-hop QA (Zhao et al., 2021; Xiong et al., 2021), and multi-hop explanations (Valentino et al., 2021; Cartuyvels et al., 2020).”
Explainable Natural Language Reasoning via Conceptual Unification (2020, methodology): “To deal with semantic drift, recent work have proposed the crowd-sourcing of explanationcentred corpora (Xie et al., 2020; Jansen et al., 2018, 2016) which can enable the identification”
Autoregressive Reasoning over Chains of Facts with Transformers (2020, methodology): “In the 2020 version of the task Df contains 9727 facts, and Dtrain qa , Dval qa , Dtest qa contain 2206, 496 and 1664 questions respectively (Xie et al., 2020).”
multiPRover: Generating Multiple Proofs for Improved Interpretability in Rule Reasoning (2021, background): “…(3) structured explanations that range from semi-structured text (Ye et al., 2020) to chain of facts (Khot et al., 2020; Jhamtani and Clark, 2020; Gontier et al., 2020) to explanation graphs (based on edges between chains of facts) (Jansen et al., 2018; Jansen and Ustalov, 2019; Xie et al., 2020).”
On the Diversity and Limits of Human Explanations (2021, background): “Xie et al. (2020) science QA similar to Jansen et al. (2018) Khot et al. (2020) question an- swering The HIT here is to write a test question that requires CHAINING two facts (a science fact and some other fact) to be combined.”
Teach Me to Explain: A Review of Datasets for Explainable NLP (2021, methodology): “…explanation graphs authors 1,680 O PEN B OOK QA (Mihaylov et al., 2018) open-book science QA 1 fact from W ORLD T REE crowd 5,957 W ORLD T REE V2 (Xie et al., 2020) science exam QA explanation graphs experts 5,100 QED (Lamm et al., 2020) reading comp. QA inference rules authors 8,991 QASC (Khot…”
ExplaGraphs: An Explanation Graph Generation Task for Structured Commonsense Reasoning (2021, background): “…et al., 2021; Zhang et al., 2020), and (3) structured explanations consisting of explanations graphs (Jansen et al., 2018; Jansen and Ustalov, 2019; Xie et al., 2020; Saha et al., 2020; Kalyanpur et al., 2020; Saha et al., 2021), chain of facts (Khot et al., 2020; Jhamtani and Clark, 2020; Inoue…”
Teach Me to Explain: A Review of Datasets for Explainable Natural Language Processing (2021, background): “The WORLDTREE datasets [57, 132] propose explaining elementary-school science questions with a combination of chains of facts and semi-structured text, termed “explanation graphs”.”