Two Birds with One Stone: Efficient Deep Learning over Mislabeled Data through Subset Selection
Summary: Deem selects a subset under label uncertainty by using losses and gradients to approximate the full gradient on soft labels. Framed as submodular NP-hard subset selection with a scalable approximation, it yields up to 10× speedups with no accuracy loss. (summarized by gpt-5-nano on Feb 09 2026)
Incoming Non-self Citations Over Time
Authors
- 1. Yuhao Deng (Beijing Institute of Technology)
- 2. Chengliang Chai (Beijing Institute of Technology)
- 3. Kaisen Jin (Beijing Institute of Technology)
- 4. Linan Zheng (University of Arizona)
- 5. Lei Cao (Massachusetts Institute of Technology; University of Arizona)
- 6. Ye Yuan (Beijing Institute of Technology)
- 7. Guoren Wang (Beijing Institute of Technology)
BibTeX Citation
@inproceedings{deng_sigmod25,
title = {{Two Birds with One Stone: Efficient Deep Learning over Mislabeled Data through Subset Selection}},
author = {Deng, Yuhao and Chai, Chengliang and Jin, Kaisen and Zheng, Linan and Cao, Lei and Yuan, Ye and Wang, Guoren},
series = {{SIGMOD} '25},
booktitle = {Proceedings of the {ACM} {SIGMOD} International Conference on Management of Data},
publisher = {Association for Computing Machinery},
doi = {10.1145/3728289},
url = {https://dl.acm.org/doi/10.1145/3728289},
year = {2025}
}Incoming Citations (Sorted by Pagerank)
Showing 3 of 3 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 10,411 | TACO: Token-Aware Context Optimization for Structured OpenStreetMap Query Generation | 2027 | SIGMOD | 4.9605343e-05 |
| 10,412 | Taxonomy Maintenance in the Wild over Evolving Scholarly Data: Reliability, Efficiency, and Cost-Effectiveness | 2027 | SIGMOD | 4.9605343e-05 |
| 10,892 | Data-efficient Online Training for Direct Alignment in LLMs | 2026 | VLDB | 4.9605343e-05 |
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 6 of 6 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 15 | How Good Are Query Optimizers, Really? | 2016 | VLDB | 0.00061463175 |
| 460 | ActiveClean: Interactive Data Cleaning For Statistical Modeling | 2016 | VLDB | 0.00017836918 |
| 1,744 | Nearest Neighbor Classifiers over Incomplete Information: From Certain Answers to Certain Predictions | 2021 | VLDB | 9.7253583e-05 |
| 3,757 | GoodCore: Data-effective and Data-efficient Machine Learning through Coreset Selection over Incomplete Data | 2023 | SIGMOD | 7.0254733e-05 |
| 6,654 | Coresets over Multiple Tables for Feature-rich and Data-efficient Machine Learning | 2023 | VLDB | 5.7038474e-05 |
| 7,667 | MisDetect: Iterative Mislabel Detection using Early Loss | 2024 | VLDB | 5.4565877e-05 |
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 7,440 | Cardinality Estimation of Approximate Substring Queries using Deep Learning | 2022 | VLDB |
| 2 | 5,153 | Monotonic Cardinality Estimation of Similarity Selection: A Deep Learning Approach | 2020 | SIGMOD |
| 3 | 8,664 | Efficient Summarization Framework for Multi-Attribute Uncertain Data | 2014 | SIGMOD |
| 4 | 3,757 | GoodCore: Data-effective and Data-efficient Machine Learning through Coreset Selection over Incomplete Data | 2023 | SIGMOD |
| 5 | 13,626 | Effective Clustering for Large Multi-Relational Graphs | 2026 | SIGMOD |
| 6 | 11,570 | Certain and Approximately Certain Models for Statistical Learning | 2024 | SIGMOD |
| 7 | 7,667 | MisDetect: Iterative Mislabel Detection using Early Loss | 2024 | VLDB |
| 8 | 1,583 | Deep Learning Models for Selectivity Estimation of Multi-Attribute Queries | 2020 | SIGMOD |
| 9 | 8,297 | Consistent and Flexible Selectivity Estimation for High-Dimensional Data | 2021 | SIGMOD |
| 10 | 9,365 | Deep Query Optimization | 2019 | SIGMOD |