mlsanityData to 14 Oct 2026

The Utility of "Even if..." Semifactual Explanation to Optimise Positive Outcomes

arXiv 2310.18937 ↗First posted 29 Oct 2023cs.AI cs.LGcited by 8

Not on the map. Why?

Abstract (arXiv)

When users receive either a positive or negative outcome from an automated system, Explainable AI (XAI) has almost exclusively focused on how to mutate negative outcomes into positive ones by crossing a decision boundary using counterfactuals (e.g., textit"If you earn 2k more, we will accept your loan application"). Here, we instead focus on textitpositive outcomes, and take the novel step of using XAI to optimise them (e.g., textit"Even if you wish to half your down-payment, we will still accept your loan application"). Explanations such as these that employ "even if..." reasoning, and do not cross a decision boundary, are known as semifactuals. To instantiate semifactuals in this context, we introduce the concept of textitGain (i.e., how much a user stands to benefit from the explanation), and consider the first causal formalisation of semifactuals. Tests on benchmark datasets show our algorithms are better at maximising gain compared to prior work, and that causality is important in the process. Most importantly however, a user study supports our main hypothesis by showing people find semifactual explanations more useful than counterfactuals when they receive the positive outcome of a loan acceptance.

Key facts. Cited by 8 papers in the map (data through 14 Oct 2026). Builds on 32 of its 60 references that are in the map. First posted on arXiv 29 Oct 2023.

Builds on

32 of 60 references.

References · 32

  1. 2016Towards Evaluating the Robustness of Neural Networkscited by 2,998
  2. 2021Convex optimizationcited by 2,515
  3. 2017Counterfactual Explanations without Opening the Black Box: Automated Decisions and the GDPRcited by 1,089
  4. 2019Explaining Machine Learning Classifiers through Diverse Counterfactual Explanationscited by 614
  5. 2018Actionable Recourse in Linear Classificationcited by 328
  6. 2018Explanations based on the Missing: Towards Contrastive Explanations with Pertinent Negativescited by 262
  7. 2018Metrics for Explainable AI: Challenges and Prospectscited by 260
  8. 2020Algorithmic Recourse: from Counterfactual Explanations to Interventionscited by 250
  9. 2019Interpretable Counterfactual Explanations Guided by Prototypescited by 223
  10. 2019Model-Agnostic Counterfactual Explanations for Consequential Decisionscited by 204
  11. 2017Fair Inference On Outcomescited by 144
  12. 2020Algorithmic recourse under imperfect causal knowledge: a probabilistic approachcited by 140
20 more
  1. 2020A survey of algorithmic recourse: definitions, formulations, solutions, and prospectscited by 118
  2. 2019The Dangers of Post-hoc Interpretability: Unjustified Counterfactual Explanationscited by 97
  3. 2022Exploring the Whole Rashomon Set of Sparse Decision Treescited by 81
  4. 2020On the Fairness of Causal Algorithmic Recoursecited by 70
  5. 2018Contrastive Explanation: A Structural-Model Approachcited by 65
  6. 2020On Generating Plausible Counterfactual and Semi-Factual Explanations for Deep Learningcited by 60
  7. 2021On the Adversarial Robustness of Causal Algorithmic Recoursecited by 56
  8. 2021Local Explanations via Necessity and Sufficiency: Unifying Theory and Practicecited by 45
  9. 2018Interpretable Credit Application Predictions With Counterfactual Explanationscited by 43
  10. 2022On the Robustness of Sparse Counterfactual Explanations to Adverse Perturbationscited by 32
  11. 2022On the Privacy Risks of Algorithmic Recoursecited by 32
  12. 2022A Rationale-Centric Framework for Human-in-the-loop Machine Learningcited by 27
  13. 2019ProPublica's COMPAS Data Revisitedcited by 24
  14. 2023Even if Explanations: Prior Work, Desiderata & Benchmarks for Semi-Factual XAIcited by 20
  15. 2021Consequence-aware Sequential Counterfactual Generationcited by 17
  16. 2022Features of Explainability: How users understand counterfactual and causal explanations for categorical and continuous features in XAIcited by 16
  17. 2022"Even if ..." -- Diverse Semifactual Explanations of Rejectcited by 11
  18. 2022Synthesizing explainable counterfactual policies for algorithmic recourse with program synthesiscited by 8
  19. 2022This changes to that : Combining causal and non-causal explanations to generate disease progression in capsule endoscopycited by 8
  20. 2022Alterfactual Explanations -- The Relevance of Irrelevance for Explaining AI Systemscited by 7

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 3; 2025: 1; 2026: 4.

8 papers.

  1. 2024Even-Ifs From If-Onlys: Are the Best Semi-Factual Explanations Found Using Counterfactuals As Guides?cited by 5
  2. 2024Even-if Explanations: Formal Foundations, Priorities and Complexitycited by 4
  3. 2024Semifactual Explanations for Reinforcement Learningcited by 1
  4. 2025Learning Counterfactual Outcomes Under Rank Preservationcited by 0
  5. 2026On the Hardness of Computing Counterfactual and Semifactual Explanations in XAIcited by 0
  6. 2026Informative Semi-Factuals for XAI: The Elaborated Explanations that People Prefercited by 0
  7. 2026Towards Verified and Targeted Explanations through Formal Methodscited by 0
  8. 2026Explanations-Driven Active Feature Acquisition for Algorithmic Recoursecited by 0