mlsanityData to 7 Oct 2026

CEBaB: Estimating the Causal Effects of Real-World Concepts on NLP Model Behavior

arXiv 2205.14140 ↗First posted 27 May 2022cs.CLcited by 54

Not on the map. Why?

Abstract (arXiv)

The increasing size and complexity of modern ML systems has improved their predictive capabilities but made their behavior harder to explain. Many techniques for model explanation have been developed in response, but we lack clear criteria for assessing these techniques. In this paper, we cast model explanation as the causal inference problem of estimating causal effects of real-world concepts on the output behavior of ML models given actual input data. We introduce CEBaB, a new benchmark dataset for assessing concept-based explanation methods in Natural Language Processing (NLP). CEBaB consists of short restaurant reviews with human-generated counterfactual reviews in which an aspect (food, noise, ambiance, service) of the dining experience was modified. Original and counterfactual reviews are annotated with multiply-validated sentiment ratings at the aspect-level and review-level. The rich structure of CEBaB allows us to go beyond input features to study the effects of abstract, real-world concepts on model behavior. We use CEBaB to compare the quality of a range of concept-based explanation methods covering different assumptions and conceptions of the problem, and we seek to establish natural metrics for comparative assessments of these methods.

Key facts. Cited by 54 papers in the map (data through 7 Oct 2026). Builds on 44 of its 70 references that are in the map. First posted on arXiv 27 May 2022.

Builds on

44 of 70 references.

References · 44

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  2. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,491
  3. 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,266
  4. 2017A Unified Approach to Interpreting Model Predictionscited by 6,488
  5. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,040
  6. 2018GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understandingcited by 4,854
  7. 2017Axiomatic Attribution for Deep Networkscited by 3,350
  8. 2013Visualizing and Understanding Convolutional Networkscited by 2,278
  9. 2019HuggingFace's Transformers: State-of-the-art Natural Language Processingcited by 2,201
  10. 2016Equality of Opportunity in Supervised Learningcited by 2,000
  11. 2019SuperGLUE: A Stickier Benchmark for General-Purpose Language Understanding Systemscited by 1,718
  12. 2016Concrete Problems in AI Safetycited by 1,582
32 more
  1. 2017Learning Important Features Through Propagating Activation Differencescited by 1,454
  2. 2017Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)cited by 1,309
  3. 2019BERT Rediscovers the Classical NLP Pipelinecited by 1,046
  4. 2015Effective Approaches to Attention-based Neural Machine Translationcited by 1,024
  5. 2019What Does BERT Look At? An Analysis of BERT's Attentioncited by 994
  6. 2014Striving for Simplicity: The All Convolutional Netcited by 983
  7. 2018A Survey Of Methods For Explaining Black Box Modelscited by 970
  8. 2020XTREME: A Massively Multilingual Multi-task Benchmark for Evaluating Cross-lingual Generalizationcited by 738
  9. 2016The Mythos of Model Interpretabilitycited by 716
  10. 2020Towards Faithfully Interpretable NLP Systems: How should we define and evaluate faithfulness?cited by 590
  11. 2019ERASER: A Benchmark to Evaluate Rationalized NLP Modelscited by 523
  12. 2017Meta-learners for Estimating Heterogeneous Treatment Effects using Machine Learningcited by 475
  13. 2018A Benchmark for Interpretability Methods in Deep Neural Networkscited by 467
  14. 2018What you can cram into a single vector: Probing sentence embeddings for linguistic propertiescited by 452
  15. 2021Dynabench: Rethinking Benchmarking in NLPcited by 447
  16. 2020Null It Out: Guarding Protected Attributes by Iterative Nullspace Projectioncited by 442
  17. 2021Causal Abstractions of Neural Networkscited by 401
  18. 2016European Union regulations on algorithmic decision-making and a "right to explanation"cited by 343
  19. 2016FastText.zip: Compressing text classification modelscited by 331
  20. 2019On Completeness-aware Concept-Based Explanations in Deep Neural Networkscited by 239
  21. 2016Layer-wise Relevance Propagation for Neural Networks with Local Renormalization Layerscited by 197
  22. 2020Evaluating Explainable AI: Which Algorithmic Explanations Help Users Predict Model Behavior?cited by 189
  23. 2020Causal Mediation Analysis for Interpreting Neural NLP: The Case of Gender Biascited by 183
  24. 2021Causal Inference in Natural Language Processing: Estimation, Prediction, Interpretation and Beyondcited by 177
  25. 2019Explaining Classifiers with Causal Concept Effect (CaCE)cited by 135
  26. 2021Causal Analysis of Syntactic Agreement Mechanisms in Neural Language Modelscited by 127
  27. 2020The Language Interpretability Tool: Extensible, Interactive Visualizations and Analysis for NLP Modelscited by 125
  28. 2019Utilizing BERT for Aspect-Based Sentiment Analysis via Constructing Auxiliary Sentencecited by 121
  29. 2020CausaLM: Causal Model Explanation Through Counterfactual Language Modelscited by 117
  30. 2020Evaluating Explanations: How much do explanations from the teacher aid students?cited by 94
  31. 2022DoCoGen: Domain Counterfactual Generation for Low Resource Domain Adaptationcited by 36
  32. 2019Discovering the Compositional Structure of Vector Representations with Role Learning Networkscited by 33

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 2; 2023: 16; 2024: 17; 2025: 9; 2026: 10.

54 papers.

Sort:

By area:

  1. 2023Finding Alignments Between Interpretable Causal Variables and Distributed Neural Representationscited by 197
  2. 2023Causal Abstraction: A Theoretical Foundation for Mechanistic Interpretabilitycited by 194
  3. 2023Causal Parrots: Large Language Models May Talk Causality But Are Not Causalcited by 146
  4. 2022Towards Faithful Model Explanation in NLP: A Surveycited by 129
  5. 2023Interpretability at Scale: Identifying Causal Mechanisms in Alpacacited by 102
  6. 2023Concept-based Explainable Artificial Intelligence: A Surveycited by 97
  7. 2024RAVEL: Evaluating Interpretability Methods on Disentangling Language Model Representationscited by 77
  8. 2025The Alternative Annotator Test for LLM-as-a-Judge: How to Statistically Justify Replacing Human Annotators with LLMscited by 72
  9. 2025MIB: A Mechanistic Interpretability Benchmarkcited by 49
  10. 2023Explore Spurious Correlations at the Concept Level in Language Models for Text Classificationcited by 48
  11. 2023Rigorously Assessing Natural Language Explanations of Neuronscited by 44
  12. 2023Faithful Explanations of Black-box NLP Models Using LLM-generated Counterfactualscited by 44
38 more
  1. 2023Interpreting Pretrained Language Models via Concept Bottleneckscited by 33
  2. 2025Inoculation Prompting: Instructing LLMs to misbehave at train-time improves test-time alignmentcited by 32
  3. 2022Causal Proxy Models for Concept-Based Model Explanationscited by 29
  4. 2024On Behalf of the Stakeholders: Trends in NLP Model Interpretability in the Era of LLMscited by 24
  5. 2023A Geometric Notion of Causal Probingcited by 23
  6. 2024Interpretable Concept-Based Memory Reasoningcited by 17
  7. 2024A Reply to Makelov et al. (2023)'s "Interpretability Illusion" Argumentscited by 15
  8. 2023A Review of the Role of Causality in Developing Trustworthy AI Systemscited by 14
  9. 2023Data Augmentations for Improved (Large) Language Model Generalizationcited by 14
  10. 2024Latent Concept-based Explanation of NLP Modelscited by 13
  11. 2024Poser: Unmasking Alignment Faking LLMs by Manipulating Their Internalscited by 13
  12. 2024Causality-inspired Discriminative Feature Learning in Triple Domains for Gait Recognitioncited by 13
  13. 2024Gumbel Counterfactual Generation From Language Modelscited by 13
  14. 2024Tuning-Free Accountable Intervention for LLM Deployment -- A Metacognitive Approachcited by 11
  15. 2025ConSim: Measuring Concept-Based Explanations' Effectiveness with Automated Simulatabilitycited by 9
  16. 2023Has It All Been Solved? Open NLP Research Questions Not Solved by Large Language Modelscited by 7
  17. 2024RATE: Causal Explainability of Reward Models with Imperfect Counterfactualscited by 6
  18. 2023Competence-Based Analysis of Language Modelscited by 4
  19. 2024Explaining Text Classifiers with Counterfactual Representationscited by 3
  20. 2024Linearly-Interpretable Concept Embedding Models for Text Analysiscited by 3
  21. 2025V-CEM: Bridging Performance and Intervenability in Concept-based Modelscited by 3
  22. 2025Interpreting Language Models Through Concept Descriptions: A Surveycited by 3
  23. 2023Explainable AI for clinical risk prediction: a survey of concepts, methods, and modalitiescited by 2
  24. 2025Nonlinear Concept Erasure: a Density Matching Approachcited by 2
  25. 2024Caformer: Rethinking Time Series Analysis from Causal Perspectivecited by 1
  26. 2024Latent Causal Probing: A Formal Perspective on Probing with Causal Models of Datacited by 1
  27. 2024Seeing Through VisualBERT: A Causal Adventure on Memetic Landscapescited by 1
  28. 2024Linking Model Intervention to Causal Interpretation in Model Explanationcited by 1
  29. 2025Enhancing the Comprehensibility of Text Explanations via Unsupervised Concept Discoverycited by 1
  30. 2026LIBERTy: A Causal Framework for Benchmarking Concept-Based Explanations of LLMs with Structural Counterfactualscited by 1
  31. 2026Consistency Training Can Entrench Misalignmentcited by 1
  32. 2023Flexible Model Interpretability through Natural Language Model Editingcited by 0
  33. 2024Position Debiasing Fine-Tuning for Causal Perception in Long-Term Dialoguecited by 0
  34. 2025CAuSE: Decoding Multimodal Classifiers using Faithful Natural Language Explanationcited by 0
  35. 2026Vector Quantized Latent Concepts: A Scalable Alternative to Clustering-Based Concept Discoverycited by 0
  36. 2026Credal Concept Bottleneck Models for Epistemic-Aleatoric Uncertainty Decompositioncited by 0
  37. 2026Causal Probing for Internal Visual Representations in Multimodal Large Language Modelscited by 0
  38. 2026Concise and Logically Consistent Conformal Sets for Neuro-Symbolic Concept-Based Modelscited by 0

50 most cited of 54 All 54 (JSON)