Explaining Black-Box Algorithms Using Probabilistic Contrastive Counterfactuals

Summary: Introduces LEWIS, a causality-based XAI framework using probabilistic counterfactuals to explain black-box decisions from input-output data. Provides provably effective local/global explanations and recourse, outperforming LIME/SHAP on real data. (summarized by gpt-5-nano on Feb 09 2026)

Paper ID: 6322
Venue: SIGMOD
Year: 2021
Pagerank: 7.8677069e-05
Overall Rank: 3,000 | 79.42%
DOI: 10.1145/3448016.3458455

Incoming Non-self Citations Over Time

Authors

1. Sainyam Galhotra (University of Massachusetts Amherst)
2. Romila Pradhan (University of California San Diego)
3. Babak Salimi (University of California San Diego)

BibTeX Citation

@inproceedings{galhotra_sigmod21,
        title = {{Explaining Black-Box Algorithms Using Probabilistic Contrastive Counterfactuals}},
        author = {Galhotra, Sainyam and Pradhan, Romila and Salimi, Babak},
        series = {{SIGMOD} '21},
        booktitle = {Proceedings of the {ACM} {SIGMOD} International Conference on Management of Data},
        publisher = {Association for Computing Machinery},
        doi = {10.1145/3448016.3458455},
        url = {https://dl.acm.org/doi/10.1145/3448016.3458455},
        year = {2021}
}

Incoming Citations (Sorted by Pagerank)

Showing 17 of 17 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank
4,876	XInsight: eXplainable Data Analysis Through The Lens of Causality	2023	SIGMOD	6.4687705e-05
5,025	HYPER: Hypothetical Reasoning With What-If and How-To Queries Using a Probabilistic Causal Approach	2022	SIGMOD	6.3974766e-05
5,054	Explainable AI: Foundations, Applications, Opportunities for Data Management Research	2022	SIGMOD	6.3843089e-05
6,079	FACET: Robust Counterfactual Explanation Analytics	2023	SIGMOD	5.9850223e-05
6,352	OTClean: Data Cleaning for Conditional Independence Violations using Optimal Transport	2024	SIGMOD	5.9042757e-05
7,192	Generating Interpretable Data-Based Explanations for Fairness Debugging using Gopher	2022	SIGMOD	5.6772818e-05
9,055	Causal DAG Summarization	2025	VLDB	5.3251649e-05
10,391	Privacy-preserving and Verifiable Causal Prescriptive Analytics	2026	SIGMOD	5.093636e-05
10,436	Causal Explanations for Disparate Trends: Where and Why?	2026	SIGMOD	5.093636e-05
10,502	Stress-Testing Causal Claims via Cardinality Repairs	2026	SIGMOD	5.093636e-05
10,709	CausalExplain: Causal Explanations of Black-box Models with Training Data Subsets	2025	SIGMOD	5.093636e-05
10,796	Understanding the Black Box: A Deep Empirical Dive into Shapley Value Approximations for Tabular Data	2025	SIGMOD	5.093636e-05
10,979	Finding Convincing Views to Endorse a Claim	2025	VLDB	5.093636e-05
11,170	Counterfactual Explanation at Will, with Zero Privacy Leakage	2024	SIGMOD	5.093636e-05
11,513	Augmenting Decision Making via Interactive What-If Analysis	2022	CIDR	5.093636e-05
11,544	CFDB: Machine Learning Model Analysis via Databases of CounterFactuals	2022	SIGMOD	5.093636e-05
13,462	Demonstration of Generating Explanations for Black-Box Algorithms Using Lewis	2021	VLDB	-

Outgoing Citations (Sorted by Pagerank)

Showing 4 of 4 cited papers.

Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.

Rank	Cited Paper	Year	Venue	Pagerank
863	Interventional Fairness : Causal Database Repair for Algorithmic Fairness	2019	SIGMOD	0.00013531835
2,374	Causal Relational Learning	2020	SIGMOD	8.6755064e-05
2,546	Bias in OLAP Queries: Detection, Explanation, and Removal (Or Think Twice About Your AVG-Query)	2018	SIGMOD	8.4340413e-05
7,122	HypDB: A Demonstration of Detecting, Explaining and Resolving Bias in OLAP queries	2018	VLDB	5.6968152e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	10,825	Explaining Black-Box Clustering Pipelines With Cluster-Explorer	2025	VLDB
2	6,637	Causal Feature Selection for Algorithmic Fairness	2022	SIGMOD
3	5,054	Explainable AI: Foundations, Applications, Opportunities for Data Management Research	2022	SIGMOD
4	1,951	Interpretable Data-Based Explanations for Fairness Debugging	2022	SIGMOD
5	11,170	Counterfactual Explanation at Will, with Zero Privacy Leakage	2024	SIGMOD
6	6,576	Toward Interpretable and Actionable Data Analysis with Explanations and Causality	2022	VLDB
7	7,907	Provenance-Enabled Explainable AI	2024	SIGMOD
8	10,709	CausalExplain: Causal Explanations of Black-box Models with Training Data Subsets	2025	SIGMOD
9	9,313	Computing Rule-Based Explanations by Leveraging Counterfactuals	2023	VLDB
10	13,462	Demonstration of Generating Explanations for Black-Box Algorithms Using Lewis	2021	VLDB