Complaint-Driven Training Data Debugging at Interactive Speeds

Summary: Rain++ enables complaint-driven debugging of training data for inference queries by ranking offending examples from complaints. Precomputation decouples cost from model size, enabling interactive ~1 ms latency for multi-million-parameter models and supporting standing/streaming queries. (summarized by gpt-5-nano on Feb 09 2026)

Paper ID: h0501b3592677ea20
Venue: SIGMOD
Year: 2022
Pagerank: 5.2364886e-05
Overall Rank: 9,019 | 39.37%
DOI: 10.1145/3514221.3517849

Incoming Non-self Citations Over Time

Authors

1. Lampros Flokas (Columbia University)
2. Weiyuan Wu (Simon Fraser University)
3. Yejia Liu (Simon Fraser University)
4. Jiannan Wang (Simon Fraser University)
5. Nakul Verma (Columbia University)
6. Eugene Wu (Columbia University)

BibTeX Citation

@inproceedings{flokas_sigmod22,
        title = {{Complaint-Driven Training Data Debugging at Interactive Speeds}},
        author = {Flokas, Lampros and Wu, Weiyuan and Liu, Yejia and Wang, Jiannan and Verma, Nakul and Wu, Eugene},
        series = {{SIGMOD} '22},
        booktitle = {Proceedings of the {ACM} {SIGMOD} International Conference on Management of Data},
        publisher = {Association for Computing Machinery},
        doi = {10.1145/3514221.3517849},
        url = {https://dl.acm.org/doi/10.1145/3514221.3517849},
        year = {2022}
}

Incoming Citations (Sorted by Pagerank)

Showing 2 of 2 citing papers.

Rank	Citing Paper	Year	Venue	Pagerank
4,982	XInsight: eXplainable Data Analysis Through The Lens of Causality	2023	SIGMOD	6.328859e-05
7,538	Automating and Optimizing Data-Centric What-If Analyses on Native Machine Learning Pipelines	2023	SIGMOD	5.4995874e-05

Outgoing Citations (Sorted by Pagerank)

Showing 11 of 11 cited papers.

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

Rank	Cited Paper	Year	Venue	Pagerank
17	Provenance Semirings	2007	PODS	0.00059752575
105	The MADlib Analytics Library or MAD Skills, the SQL	2012	VLDB	0.00033638251
189	Scorpion: Explaining Away Outliers in Aggregate Queries	2013	VLDB	0.00025840026
819	Provenance for Aggregate Queries	2011	PODS	0.00013666629
1,051	Tiresias: The Database Oracle for How-To Queries	2012	SIGMOD	0.00012279821
1,308	Automating Large-Scale Data Quality Verification	2018	VLDB	0.0001107886
2,160	DIFF: A Relational Interface for Large-Scale Data Explanation	2019	VLDB	8.9364035e-05
2,598	Complaint-driven Training Data Debugging for Query 2.0	2020	SIGMOD	8.2385793e-05
2,686	Data X-Ray: A Diagnostic Tool for Data Errors	2015	SIGMOD	8.1308928e-05
4,195	PrIU: A Provenance-Based Approach for Incrementally Updating Regression Models	2020	SIGMOD	6.7422305e-05
4,739	Going Beyond Provenance: Explaining Query Answers with Pattern-based Counterbalances	2019	SIGMOD	6.4441962e-05

Semantically Similar Papers

#	Overall Rank	Paper	Year	Venue
1	7,363	PerfGuard: Deploying ML-for-Systems without Performance Regressions, Almost!	2021	VLDB
2	6,844	Serving Deep Learning Models with Deduplication from Relational Databases	2022	VLDB
3	5,272	InferDB: In-Database Machine Learning Inference Using Indexes	2024	VLDB
4	560	Plan-Structured Deep Neural Network Models for Query Performance Prediction	2019	VLDB
5	318	DeepDB: Learn from Data, not from Queries!	2020	VLDB
6	5,158	Facilitating SQL Query Composition and Analysis	2020	SIGMOD
7	281	Accelerating Machine Learning Inference with Probabilistic Predicates	2018	SIGMOD
8	7,236	RCRank: Multimodal Ranking of Root Causes of Slow Queries in Cloud Database Systems	2025	VLDB
9	5,247	Enabling SQL-based Training Data Debugging for Federated Learning	2022	VLDB
10	2,598	Complaint-driven Training Data Debugging for Query 2.0	2020	SIGMOD