MLINSPECT: A Data Distribution Debugger for Machine Learning Pipelines
Summary: MLINSPECT: a data distribution debugger for ML pipelines, using lightweight lineage-based inspection to pinpoint distribution bugs in preprocessing. It operates on declarative estimator/transformer abstractions, handles relational and matrix data, and requires no manual code instrumentation. (summarized by gpt-5-nano on Feb 09 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Stefan Grafberger (University of Amsterdam)
- 2. Shubha Guha (University of Amsterdam)
- 3. Julia Stoyanovich (New York University)
- 4. Sebastian Schelter (University of Amsterdam)
BibTeX Citation
@inproceedings{grafberger_sigmod21,
title = {{MLINSPECT: A Data Distribution Debugger for Machine Learning Pipelines}},
author = {Grafberger, Stefan and Guha, Shubha and Stoyanovich, Julia and Schelter, Sebastian},
series = {{SIGMOD} '21},
booktitle = {Proceedings of the {ACM} {SIGMOD} International Conference on Management of Data},
publisher = {Association for Computing Machinery},
doi = {10.1145/3448016.3452759},
url = {https://dl.acm.org/doi/10.1145/3448016.3452759},
year = {2021}
}Incoming Citations (Sorted by Pagerank)
Showing 9 of 9 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 1,941 | Interpretable Data-Based Explanations for Fairness Debugging | 2022 | SIGMOD | 9.3297671e-05 |
| 4,327 | Query Refinement for Diversity Constraint Satisfaction | 2024 | VLDB | 6.6632438e-05 |
| 7,538 | Automating and Optimizing Data-Centric What-If Analyses on Native Machine Learning Pipelines | 2023 | SIGMOD | 5.4995874e-05 |
| 7,629 | mlwhatif: What If You Could Stop Re-Implementing Your Machine Learning Pipeline Analyses Over and Over? | 2023 | VLDB | 5.4803665e-05 |
| 8,411 | SHiFT: An Efficient, Flexible Search Engine for Transfer Learning | 2023 | VLDB | 5.336291e-05 |
| 9,417 | Towards Observability for Production Machine Learning Pipelines | 2022 | VLDB | 5.1803615e-05 |
| 10,803 | Efficient Query Repair for Aggregate Constraints | 2026 | VLDB | 4.9793485e-05 |
| 11,410 | APEX-DAG: Library and Language independent Pipeline EXtraction | 2025 | VLDB | 4.9793485e-05 |
| 11,821 | Towards Observability for Machine Learning Pipelines | 2022 | CIDR | 4.9793485e-05 |
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 2 of 2 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 1,162 | Responsible Data Management | 2020 | VLDB | 0.00011753159 |
| 6,323 | Lightweight Inspection of Data Preprocessing in Native Machine Learning Pipelines | 2021 | CIDR | 5.8138422e-05 |
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 9,675 | Debugging Large-Scale Data Science Pipelines using Dagger | 2020 | VLDB |
| 2 | 11,407 | mlidea: Interactively Improving ML Data Preparation Code via “Shadow Pipelines” | 2025 | VLDB |
| 3 | 11,410 | APEX-DAG: Library and Language independent Pipeline EXtraction | 2025 | VLDB |
| 4 | 10,855 | PipeLens: Identifying Interventions for Resolving Malfunctioning Data Science Pipelines | 2026 | VLDB |
| 5 | 7,538 | Automating and Optimizing Data-Centric What-If Analyses on Native Machine Learning Pipelines | 2023 | SIGMOD |
| 6 | 9,417 | Towards Observability for Production Machine Learning Pipelines | 2022 | VLDB |
| 7 | 7,629 | mlwhatif: What If You Could Stop Re-Implementing Your Machine Learning Pipeline Analyses Over and Over? | 2023 | VLDB |
| 8 | 11,821 | Towards Observability for Machine Learning Pipelines | 2022 | CIDR |
| 9 | 11,671 | Reconstructing and Querying ML Pipeline Intermediates | 2023 | CIDR |
| 10 | 6,323 | Lightweight Inspection of Data Preprocessing in Native Machine Learning Pipelines | 2021 | CIDR |