IDAP++: Advancing Divergence-Aware Pruning with Joint Filter and Layer Optimization
Summary: IDAP++ proposes divergence-aware neural compression with a unified information-flow metric. It extends pruning from filters to whole layers via joint filter/layer optimization, enabling architecture-agnostic compression across CNNs, transformers, and hybrids. (summarized by gpt-5-mini on Apr 11 2026)
Incoming Non-self Citations Over Time
No non-self incoming citations found for this paper in this database.
Authors
- 1. Aleksei Samarin (ITMO University; Wayy LLC)
- 2. Artem Nazarenko (ITMO University; Wayy LLC)
- 3. Egor Kotenko (Saint Petersburg State University; Wayy LLC)
- 4. Alexander Savelev (ITMO University; Wayy LLC)
- 5. Aleksei Toropov (ITMO University; Wayy LLC)
- 6. Alexandr Motyko (Wayy LLC)
- 7. Valentin Malykh (ITMO University; Wayy LLC)
BibTeX Citation
@inproceedings{samarin_sigmod26,
title = {{IDAP++: Advancing Divergence-Aware Pruning with Joint Filter and Layer Optimization}},
author = {Samarin, Aleksei and Nazarenko, Artem and Kotenko, Egor and Savelev, Alexander and Toropov, Aleksei and Motyko, Alexandr and Malykh, Valentin},
series = {{SIGMOD} '26},
booktitle = {Proceedings of the {ACM} {SIGMOD} International Conference on Management of Data},
publisher = {Association for Computing Machinery},
doi = {10.1145/3786659},
url = {https://dl.acm.org/doi/10.1145/3786659},
year = {2026}
}Incoming Citations (Sorted by Pagerank)
Showing 0 of 0 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 3 of 3 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 154 | Neo: A Learned Query Optimizer | 2019 | VLDB | 0.00028726181 |
| 323 | DeepDB: Learn from Data, not from Queries! | 2020 | VLDB | 0.00021264788 |
| 447 | ALEX: An Updatable Adaptive Learned Index | 2020 | SIGMOD | 0.00018322593 |
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 10,386 | NeurStore: Efficient In-database Deep Learning Model Management System | 2026 | SIGMOD |
| 2 | 10,589 | Morphing-based Compression for Data-centric ML Pipelines | 2026 | VLDB |
| 3 | 11,553 | SparRL: Graph Sparsification via Deep Reinforcement Learning | 2022 | SIGMOD |
| 4 | 8,794 | AWARE: Workload-aware, Redundancy-exploiting Linear Algebra | 2023 | SIGMOD |
| 5 | 1,644 | Compressed Linear Algebra for Large-Scale Machine Learning | 2016 | VLDB |
| 6 | 13,354 | STile: Searching Hybrid Sparse Formats for Sparse Deep Learning Operators Automatically | 2024 | SIGMOD |
| 7 | 2,386 | Accelerating Large Scale Real-Time GNN Inference using Channel Pruning | 2021 | VLDB |
| 8 | 3,092 | DeepSqueeze: Deep Semantic Compression for Tabular Data | 2020 | SIGMOD |
| 9 | 8,908 | Privacy and Accuracy-Aware AI/ML Model Deduplication | 2025 | SIGMOD |
| 10 | 10,913 | Inference-friendly Graph Compression for Graph Neural Networks | 2025 | VLDB |