RITA: Group Attention is All You Need for Timeseries Analytics
Summary: RITA uses group attention to scale time-series transformers by clustering similar series into few groups and attending groupwise. Dynamic scheduler adapts group count and batch size to meet quality guarantees, delivering up to 63× speedups with accuracy. (summarized by gpt-5-nano on Feb 09 2026)
Incoming Non-self Citations Over Time
No non-self incoming citations found for this paper in this database.
Authors
- 1. Jiaming Liang (University of Pennsylvania)
- 2. Lei Cao (Massachusetts Institute of Technology; University of Arizona)
- 3. Samuel Madden (Massachusetts Institute of Technology)
- 4. Zachary Ives (University of Pennsylvania)
- 5. Guoliang Li (Tsinghua University)
BibTeX Citation
@inproceedings{liang_sigmod24,
title = {{RITA: Group Attention is All You Need for Timeseries Analytics}},
author = {Liang, Jiaming and Cao, Lei and Madden, Samuel and Ives, Zachary and Li, Guoliang},
series = {{SIGMOD} '24},
booktitle = {Proceedings of the {ACM} {SIGMOD} International Conference on Management of Data},
publisher = {Association for Computing Machinery},
doi = {10.1145/3639317},
url = {https://dl.acm.org/doi/10.1145/3639317},
year = {2024}
}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 5 of 5 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 286 | Milvus: A Purpose-Built Vector Data Management System | 2021 | SIGMOD | 0.00022357911 |
| 371 | Fast Similarity Search in the Presence of Noise, Scaling, and Translation in Time-Series Databases | 1995 | VLDB | 0.00019869565 |
| 3,861 | RobustPeriod: Robust Time-Frequency Mining for Multiple Periodicity Detection | 2021 | SIGMOD | 7.0666117e-05 |
| 3,869 | Smile: A System to Support Machine Learning on EEG Data at Scale | 2019 | VLDB | 7.0609879e-05 |
| 4,234 | GRAIL: Efficient Time-Series Representation Learning | 2019 | VLDB | 6.8171563e-05 |
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 4,590 | Entity Resolution with Hierarchical Graph Attention Networks | 2022 | SIGMOD |
| 2 | 5,038 | Time2Feat: Learning Interpretable Representations for Multivariate Time Series Clustering | 2023 | VLDB |
| 3 | 13,320 | A Memory Guided Transformer for Time Series Forecasting | 2025 | VLDB |
| 4 | 10,271 | MUFASA: Fast and Accurate Multivariate Time-Series Clustering | 2026 | SIGMOD |
| 5 | 810 | Similarity-Based Queries for Time Series Data | 1997 | SIGMOD |
| 6 | 11,266 | DARKER: Efficient Transformer with Data-driven Attention Mechanism for Time Series | 2024 | VLDB |
| 7 | 6,168 | GRETA: Graph-based Real-time Event Trend Aggregation | 2018 | VLDB |
| 8 | 4,234 | GRAIL: Efficient Time-Series Representation Learning | 2019 | VLDB |
| 9 | 11,921 | DITA: A Distributed In-Memory Trajectory Analytics System | 2018 | SIGMOD |
| 10 | 2,416 | DITA: Distributed In-Memory Trajectory Analytics | 2018 | SIGMOD |