• Corpus ID: 232035689

Teach Me to Explain: A Review of Datasets for Explainable NLP

@article{Wiegreffe2021TeachMT,
  title={Teach Me to Explain: A Review of Datasets for Explainable NLP},
  author={Sarah Wiegreffe and Ana Marasovi{\'c}},
  journal={ArXiv},
  year={2021},
  volume={abs/2102.12060}
}
Explainable NLP (EXNLP) has increasingly focused on collecting humanannotated explanations. These explanations are used downstream in three ways: as data augmentation to improve performance on a predictive task, as a loss signal to train models to produce explanations for their predictions, and as a means to evaluate the quality of model-generated explanations. In this review, we identify three predominant classes of explanations (highlights, free-text, and structured), organize the literature… 

Figures and Tables from this paper

Double Trouble: How to not explain a text classifier's decisions using counterfactuals synthesized by masked language models?
TLDR
A rigorous evaluation of IM explanations found them to be consistently more biased, less accurate, and less plausible than those derived from simply deleting a word.
On the Diversity and Limits of Human Explanations
TLDR
Inspired by prior work in psychology and cognitive sciences, existing human explanations in NLP are group into three categories: proximal mechanism, evidence, and procedure, which differ in nature and have implications for the resultant explanations.
Do Natural Language Explanations Represent Valid Logical Arguments? Verifying Entailment in Explainable NLI Gold Standards
TLDR
A systematic annotation methodology, named Explanation Entailment Verification (EEV), is proposed, to quantify the logical validity of human-annotated explanations, and confirms that the inferential properties of explanations are still poorly formalised and understood.
Hybrid Autoregressive Inference for Scalable Multi-hop Explanation Regeneration
TLDR
This paper presents SCAR (for Scalable Autoregressive Inference), a hybrid framework that iteratively combines a Transformer-based bi-encoder with a sparse model of explanatory power, designed to leverage explicit inference patterns in the explanations.
Two Instances of Interpretable Neural Network for Universal Approximations
This paper proposes two bottom-up interpretable neural network (NN) constructions for universal approximation, namely Triangularly-constructed NN (TNN) and Semi-Quantized Activation NN (SQANN). The
CoAuthor: Designing a Human-AI Collaborative Writing Dataset for Exploring Language Model Capabilities
TLDR
It is argued that by curating and analyzing large interaction datasets, the HCI community can foster more incisive examinations of LMs’ generative capabilities, and presents CoAuthor, a dataset designed for revealing GPT-3’s capabilities in assisting creative and argumentative writing.
Diagnosing AI Explanation Methods with Folk Concepts of Behavior
When explaining AI behavior to humans, how is the communicated information being comprehended by the human explainee, and does it match what the explanation attempted to communicate? When can we say
Classification of Goods Using Text Descriptions With Sentences Retrieval
TLDR
A decision model based on KoELECTRA that suggests the most likely heading and subheadings of the HS code has an accuracy of 95.5% and implies algorithms may reduce the time and effort taken by customs officers substantially by assisting this seemingly challenging HS code classification task.
ExplaGraphs: An Explanation Graph Generation Task for Structured Commonsense Reasoning
TLDR
This work presents EXPLAGRAPHS, a new generative and structured commonsense-reasoning task (and an associated dataset) of explanation graph generation for stance prediction, and proposes a multi-level evaluation framework that checks for the structural and semantic correctness of the generated graphs and their degree of match with ground-truth graphs.
Explainable Machine Learning with Prior Knowledge: An Overview
TLDR
This survey presents an overview of integrating prior knowledge into machine learning systems in order to improve explainability and presents a categorization of current research into three main categories which either integrate knowledge into the machine learning pipeline, into the explainability method or derive knowledge from explanations.
...
1
2
3
...

References

SHOWING 1-10 OF 205 REFERENCES
QED: A Framework and Dataset for Explanations in Question Answering
TLDR
A large user study is described showing that the presence of QED explanations significantly improves the ability of untrained raters to spot errors made by a strong neural QA baseline.
Evaluating and Characterizing Human Rationales
TLDR
Analysis of a variety of datasets and models finds that human rationales do not necessarily perform well on automated metrics, and proposes improved metrics to account for model-dependent baseline performance and two methods to further characterize rationale quality.
GLUCOSE: GeneraLized and COntextualized Story Explanations
TLDR
This paper presents a platform for effectively crowdsourcing GLUCOSE data at scale, which uses semi-structured templates to elicit causal explanations and collects 440K specific statements and general rules that capture implicit commonsense knowledge about everyday situations.
Learning to Explain: Datasets and Models for Identifying Valid Reasoning Chains in Multihop Question-Answering
TLDR
A delexicalized chain representation in which repeated noun phrases are replaced by variables, thus turning them into generalized reasoning chains is explored, finding that generalized chains maintain performance while also being more robust to certain perturbations.
QASC: A Dataset for Question Answering via Sentence Composition
TLDR
This work presents a multi-hop reasoning dataset, Question Answering via Sentence Composition (QASC), that requires retrieving facts from a large corpus and composing them to answer a multiple-choice question, and presents a two-step approach to mitigate the retrieval challenges.
Are We Modeling the Task or the Annotator? An Investigation of Annotator Bias in Natural Language Understanding Datasets
TLDR
It is shown that model performance improves when training with annotator identifiers as features, and that models are able to recognize the most productive annotators and that often models do not generalize well to examples from annotators that did not contribute to the training set.
Evaluating Explanation Without Ground Truth in Interpretable Machine Learning
TLDR
To benchmark the evaluation in IML, this article rigorously defines the problem of evaluating explanations, and systematically review the existing efforts from state-of-the-arts, and summarizes three general aspects of explanation with formal definitions.
Explain Yourself! Leveraging Language Models for Commonsense Reasoning
TLDR
This work collects human explanations for commonsense reasoning in the form of natural language sequences and highlighted annotations in a new dataset called Common Sense Explanations to train language models to automatically generate explanations that can be used during training and inference in a novel Commonsense Auto-Generated Explanation framework.
From Recognition to Cognition: Visual Commonsense Reasoning
TLDR
To move towards cognition-level understanding, a new reasoning engine is presented, Recognition to Cognition Networks (R2C), that models the necessary layered inferences for grounding, contextualization, and reasoning.
Inferring Which Medical Treatments Work from Reports of Clinical Trials
TLDR
A new task and corpus for inferring reported findings from a full-text article describing randomized controlled trials (RCT) with respect to a given intervention, comparator, and outcome of interest and results using a suite of baseline models demonstrate the difficulty of the task.
...
1
2
3
4
5
...