Research map / Learning theory and algorithms / Prediction and decision losses
Counterfactual explanations and recourse
Generate actionable counterfactuals and recourse for decisions of trained models.
46 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2019–2026) · an approach in Prediction and decision losses, keyword cluster “explanations · shapley · attribution”
State History Tried Checklist Preprints Who Export
Where it stands
Not enough benchmark data Fewer than four papers report the same benchmark, so there is no record to follow.
11 papers in 2023–24 and 13 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic held steady, from 2.1% to 2.3%.
How this is decided
A benchmark is tracked when at least four papers of this approach report their own method on the same dataset and metric in at least two different years. Its record is the best number reported up to 2024 compared with the best reported in 2025–26: still improving means at least 1% better, flat means at most 0.2% better with at least two papers from 2025–26 trying. Numbers on a different scale from the rest (a fraction among percentages) and far outliers are left out. Results come from different papers with different setups, so they are signals, not head-to-head comparisons. The share of papers is counted at ICML, NeurIPS, ICLR, CVPR and AAAI.
How it got here
Most cited paper of each year
- 2019 · Abduction-Based Explanations for Machine Learning Models (AAAI 2019 · 196 citations)
- 2020 · Beyond Individualized Recourse: Interpretable and Interactive Summaries of Actionable Recourses (NeurIPS 2020 · 32 citations)
- 2021 · Ordered Counterfactual Explanation by Mixed-Integer Linear Optimization (AAAI 2021 · 29 citations)
- 2022 · FOCUS: Flexible Optimizable Counterfactual Explanations for Tree Ensembles (AAAI 2022 · 55 citations)
- 2023 · Theoretical Behavior of XAI Methods in the Presence of Suppressor Variables (ICML 2023 · 8 citations)
- 2024 · Delivering Inflated Explanations (AAAI 2024 · 11 citations)
- 2025 · BoolXAI: Explainable AI Using Expressive Boolean Formulas (AAAI 2025 · 4 citations)
- 2026 · GLANCE: Global Actions in a Nutshell for Counterfactual Explainability (AAAI 2026 · 3 citations)
Earliest papers here
- Abduction-Based Explanations for Machine Learning Models (AAAI 2019 · 196 citations)
- Fairwashing: the risk of rationalization (ICML 2019 · 49 citations)
- Beyond Individualized Recourse: Interpretable and Interactive Summaries of Actionable Recourses (NeurIPS 2020 · 32 citations)
What has been tried
35 papers here report 299 measured changes (a component added, removed or swapped, compared with a baseline): helped 60%, no clear effect 8%, hurt 24%, mixed 7%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| Group t1 through t5 into one state group | Five-switch railway example · Shapley responsibility | t_i: 0.0238 -> 1/6; s1: 0.3571 -> 1/6; s2: 0.5238 -> 2/3 | Mixed | Backward Responsibility in Transition Systems Using General Power Indices (AAAI 2024 · 6 citations) |
| Optimize distance jointly with likelihood (α = 0.1) and relax the plausibility constraint rather than use the median likelihood threshold | GMSC · Similarity (MAD distance) | 10.991793505652852 -> 5.871975675449956 | Helped | Generating Likely Counterfactuals Using Sum-Product Networks (ICLR 2025 · 0 citations) |
| Use T-CREx with τ=0.99 instead of τ=0.9 | Nine binary classification datasets · accuracy | Helped | Counterfactual Metarules for Local and Global Recourse (ICML 2024 · 0 citations) |
What reviewers will expect
Datasets and baselines named in the 7 papers of this approach from 2025–2026 that have a full-text summary. Leaving out the common ones is a frequent review question.
Datasets
- Parkinsons · 2 of 7 papers
- German Credit · 2 of 7 papers
Baselines
- PROPLACE · 2 of 7 papers
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 24 papers; from OpenAlex and the papers' affiliations).
- National University of Singapore · 2 papers
- Carnegie Mellon University · 2 papers
- University of Toronto · 2 papers
- J.P. Morgan Chase · 2 papers
Use it
Download BibTeX of 34 key papersOpen the topic in the interactive map
Get alerts. Describe your idea in scime atlas and save it (free scime account): you get a weekly email when new preprints land close to it.
Related approaches
Same keyword cluster
- Provable post-hoc explanations (92 papers)
- Shapley value attribution (63 papers)
Elsewhere in the topic
- Adaptive testing and experiments (201 papers)
- Active learning and subset selection (199 papers)
- Distributionally robust optimization (184 papers)
- Data attribution and recourse (160 papers)
- Theory of neural network sensitivity (143 papers)
- Ranking models from comparisons (127 papers)
- Determinantal point process inference (122 papers)
- Conformal prediction sets (116 papers)
Benchmark numbers, ablations and baseline numbers are extracted by AI from each paper's full text and can be wrong; the paper is the authority. Data: results index of 2026-11-02, citation counts from OpenAlex (2026-10-09).