Learning to Explain: Datasets and Models for Identifying Valid Reasoning Chains in Multihop Question-Answering

By Harsh Jhamtani, Peter Clark. 2020

81 citations recorded by the source corpus.

Despite the rapid progress in multihop question-answering (QA), models still have trouble explaining why an answer is correct, with limited explanation training data available to learn from. To address this, we introduce three explanation datasets in which explanations formed from corpus facts are annotated. Our first dataset, eQASC, contains over 98K explanation annotations for the multihop question answering dataset QASC, and is the first that annotates multiple candidate explanations for each answer. The second dataset eQASC-perturbed is constructed by crowd-sourcing perturbations (while preserving their validity) of a subset of explanations in QASC, to test consistency and generalization of explanation prediction models. The third dataset eOBQA is constructed by adding explanation annotations to the OBQA dataset to test generalization of models trained on eQASC. We show that this data can be used to significantly improve explanation quality (+14% absolute F1 over a strong retrieval baseline) using a BERT-based classifier, but still behind the upper bound, offering a new challenge for future research. We also explore a delexicalized chain representation in which repeated noun phrases are replaced by variables, thus turning them into generalized reasoning chains (for example: "X is a Y" AND "Y has Z" IMPLIES "X has Z"). We find that generalized chains maintain performance while also being more robust to certain perturbations.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Natural Language Processing, Prompting & Reasoning

How researchers see this paper

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

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Natural Language Processing

In Prompting & Reasoning

Citations in the corpus

Faithful Reasoning Using Large Language Models (2022, methodology): “Jhamtani and Clark (2020) develop datasets of valid and invalid reasoning traces for grade school science questions.”
Selection-Inference: Exploiting Large Language Models for Interpretable Logical Reasoning (2022, background): “1a); 2) approaches that encourage LLMs to produce reasoning explicitly, but all reasoning steps are produced in one generative step (Cobbe et al., 2021; Dalvi et al., 2021; Jhamtani and Clark, 2020; Nye et al., 2021a; Wei et al., 2022; Zelikman et al., 2022) (e.g. Fig.”
Learn to Explain: Multimodal Reasoning via Thought Chains for Science Question Answering (2022, background): “To address this issue, other science datasets annotate the explanations, but they are restricted to the textual only modality and limited to small data scales [13, 6, 37] or a small set of topics [20, 14].”
ProofWriter: Generating Implications, Proofs, and Abductive Statements over Natural Language (2020, background): “erate human-style justifications, which again are typically supporting evidence rather than a fullyformed line of reasoning, and without explicit reasoning rules (Camburu et al., 2018; Jhamtani and Clark, 2020; Inoue et al., 2020).”
LAMBADA: Backward Chaining for Automated Reasoning in Natural Language (2022, unknown): “Verifiers: To improve CoT, some works train a verifier using chain-level labels: the verifier takes a reasoning chain produced by the model as input and judges the quality of the chain (Cobbe et al., 2021; Shen et al., 2021; Jhamtani and Clark, 2020; Zelikman et al., 2022).”
Entailment Tree Explanations via Iterative Retrieval-Generation Reasoner (2022, background): “In comparison, other explanation approaches such as retrieval of passages (rationales) (DeYoung et al., 2020) or multi-hop reasoning (chaining) (Jhamtani and Clark, 2020) are less expressive than entailment trees, which are comprised of multi-premise textual entailment steps.”
multiPRover: Generating Multiple Proofs for Improved Interpretability in Rule Reasoning (2021, background): “…constrained to the input, and (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…”
Case-Based Abductive Natural Language Inference (2020, background): “, 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).”
Teach Me to Explain: A Review of Datasets for Explainable NLP (2021, methodology): “(2019a) apply a two-stage methodology to explanation collection, by either extracting explanation candidates automatically (Jhamtani and Clark, 2020) or having crowdworkers write them (Zhang et al.”
ExplaGraphs: An Explanation Graph Generation Task for Structured Commonsense Reasoning (2021, background): “…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 et al., 2020; Geva et al., 2021) or semi-structured text (Ye et al., 2020).”
Teach Me to Explain: A Review of Datasets for Explainable Natural Language Processing (2021, methodology): “In 2012 IEEE 12th International Conference on Data Mining, 2012.748 URL https://ieeexplore.ieee.org/document/6413815.749 [78] Todor Mihaylov, Peter Clark, Tushar Khot, and Ashish Sabharwal.”