mlsanityData to 7 Oct 2026

On the Diversity and Limits of Human Explanations

arXiv 2106.11988 ↗First posted 22 Jun 2021cs.CL cs.AI cs.CYcited by 30

Not on the map. Why?

Abstract (arXiv)

A growing effort in NLP aims to build datasets of human explanations. However, the term explanation encompasses a broad range of notions, each with different properties and ramifications. Our goal is to provide an overview of diverse types of explanations and human limitations, and discuss implications for collecting and using explanations in NLP. Inspired by prior work in psychology and cognitive sciences, we group existing human explanations in NLP into three categories: proximal mechanism, evidence, and procedure. These three types differ in nature and have implications for the resultant explanations. For instance, procedure is not considered explanations in psychology and connects with a rich body of work on learning from instructions. The diversity of explanations is further evidenced by proxy questions that are needed for annotators to interpret and answer open-ended why questions. Finally, explanations may require different, often deeper, understandings than predictions, which casts doubt on whether humans can provide useful explanations in some tasks.

Key facts. Cited by 30 papers in the map (data through 7 Oct 2026). Builds on 34 of its 73 references that are in the map. First posted on arXiv 22 Jun 2021.

Builds on

34 of 73 references.

References · 34

  1. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,686
  2. 2018FEVER: a large-scale dataset for Fact Extraction and VERificationcited by 1,362
  3. 2021Did Aristotle Use a Laptop? A Question Answering Benchmark with Implicit Reasoning Strategiescited by 736
  4. 2017Program Induction by Rationale Generation : Learning to Solve and Explain Algebraic Word Problemscited by 642
  5. 2020Fact or Fiction: Verifying Scientific Claimscited by 546
  6. 2019ERASER: A Benchmark to Evaluate Rationalized NLP Modelscited by 523
  7. 2018e-SNLI: Natural Language Inference with Natural Language Explanationscited by 468
  8. 2019Social Bias Frames: Reasoning about Social and Power Implications of Languagecited by 377
  9. 2019Explain Yourself! Leveraging Language Models for Commonsense Reasoningcited by 367
  10. 2019QASC: A Dataset for Question Answering via Sentence Compositioncited by 248
  11. 2018Textual Explanations for Self-Driving Vehiclescited by 226
  12. 2020Explainable Automated Fact-Checking for Public Health Claimscited by 209
22 more
  1. 2018Multimodal Explanations: Justifying Decisions and Pointing to the Evidencecited by 179
  2. 2021Teach Me to Explain: A Review of Datasets for Explainable Natural Language Processingcited by 126
  3. 2020"Why is 'Chicago' deceptive?" Towards Building Model-Driven Tutorials for Humanscited by 86
  4. 2019Automated Rationale Generation: A Technique for Explainable AI and its Effects on Human Perceptionscited by 81
  5. 2019A Richly Annotated Corpus for Different Tasks in Automated Fact-Checkingcited by 74
  6. 2019Inferring Which Medical Treatments Work from Reports of Clinical Trialscited by 63
  7. 2020What is More Likely to Happen Next? Video-and-Language Future Event Predictioncited by 61
  8. 2018WorldTree: A Corpus of Explanation Graphs for Elementary Science Questions supporting Multi-Hop Inferencecited by 60
  9. 2018Training Classifiers with Natural Language Explanationscited by 58
  10. 2020QED: A Framework and Dataset for Explanations in Question Answeringcited by 55
  11. 2018VQA-E: Explaining, Elaborating, and Enhancing Your Answers for Visual Questionscited by 50
  12. 2020Learning to Explain: Datasets and Models for Identifying Valid Reasoning Chains in Multihop Question-Answeringcited by 49
  13. 2020WinoWhy: A Deep Diagnosis of Essential Commonsense Knowledge for Answering Winograd Schema Challengecited by 43
  14. 2019R4C: A Benchmark for Evaluating RC Systems to Get the Right Answer for the Right Reasoncited by 36
  15. 2020Evaluating and Characterizing Human Rationalescited by 36
  16. 2020e-SNLI-VE: Corrected Visual-Textual Entailment with Natural Language Explanationscited by 30
  17. 2020Learning to Rationalize for Nonmonotonic Reasoning with Distant Supervisioncited by 27
  18. 2020Teaching Machine Comprehension with Compositional Explanationscited by 22
  19. 2014The effect of wording on message propagation: Topic- and author-controlled natural experiments on Twittercited by 21
  20. 2019Learning from Explanations with Neural Execution Treecited by 18
  21. 2020ESPRIT: Explaining Solutions to Physical Reasoning Taskscited by 16
  22. 2018Extractive Adversarial Networks: High-Recall Explanations for Identifying Personal Attacks in Social Media Postscited by 9

Built on by

Cited by year

Citing papers per year: 2021: 1; 2022: 10; 2023: 7; 2024: 5; 2025: 7.

30 papers.

Sort:

By area:

  1. 2021Reframing Human-AI Collaboration for Generating Free-Text Explanationscited by 125
  2. 2022Do Androids Laugh at Electric Sheep? Humor "Understanding" Benchmarks from The New Yorker Caption Contestcited by 98
  3. 2023In Search of Verifiability: Explanations Rarely Enable Complementary Performance in AI-Advised Decision Makingcited by 50
  4. 2023Are Machine Rationales (Not) Useful to Humans? Measuring and Improving Human Utility of Free-Text Rationalescited by 37
  5. 2022Going Beyond XAI: A Systematic Survey for Explanation-Guided Learningcited by 35
  6. 2024Mechanistic?cited by 25
  7. 2024Properties and Challenges of LLM-Generated Explanationscited by 24
  8. 2023Ecologically Valid Explanations for Label Variation in NLIcited by 16
  9. 2022A survey on improving NLP models with human explanationscited by 15
  10. 2023From Dissonance to Insights: Dissecting Disagreements in Rationale Construction for Case Outcome Classificationcited by 14
  11. 2022Investigating the Benefits of Free-Form Rationalescited by 13
  12. 2022Saliency Map Verbalization: Comparing Feature Importance Representations from Model-free and Instruction-based Methodscited by 13
18 more
  1. 2024Enhancing Systematic Decompositional Natural Language Inference Using Informal Logiccited by 12
  2. 2023FLamE: Few-shot Learning from Natural Language Explanationscited by 10
  3. 2023Characterizing Large Language Models as Rationalizers of Knowledge-intensive Taskscited by 9
  4. 2024Cross-Refine: Improving Natural Language Explanation Generation by Learning in Tandemcited by 8
  5. 2022ER-Test: Evaluating Explanation Regularization Methods for Language Modelscited by 6
  6. 2023Understanding and Predicting Human Label Variation in Natural Language Inference through Explanationcited by 5
  7. 2025Using Large Language Models to Categorize Strategic Situations and Decipher Motivations Behind Human Behaviorscited by 5
  8. 2025Agree, Disagree, Explain: Decomposing Human Label Variation in NLI through the Lens of Explanationscited by 5
  9. 2022Explanation Regeneration via Information Bottleneckcited by 4
  10. 2025LiTEx: A Linguistic Taxonomy of Explanations for Understanding Within-Label Variation in Natural Language Inferencecited by 4
  11. 2025Prompting as Scientific Inquirycited by 2
  12. 2022Scientific Explanation and Natural Language: A Unified Epistemological-Linguistic Perspective for Explainable AIcited by 1
  13. 2022On Faithfulness and Coherence of Language Explanations for Recommendation Systemscited by 1
  14. 2022Truthful Meta-Explanations for Local Interpretability of Machine Learning Modelscited by 1
  15. 2024A look under the hood of the Interactive Deep Learning Enterprise (No-IDLE)cited by 1
  16. 2025Complete Evidence Extraction with Model Ensembles: A Case Study on Medical Codingcited by 1
  17. 2025Emerging categories in scientific explanationscited by 0
  18. 2025Can LLM-Generated Textual Explanations Enhance Model Classification Performance? An Empirical Studycited by 0