SIMformer: Single-Layer Vanilla Transformer Can Learn Free-Space Trajectory Similarity
Summary: SIMformer shows that a single-layer vanilla Transformer can learn free-space trajectory similarity without triplet supervision or auxiliary information. Tailored non-Euclidean representation similarities mitigate dimensionality collapse while improving accuracy, speed, and scalability over prior embedding methods. (summarized by gpt-5.6-luna on Jul 24 2026)
Incoming Non-self Citations Over Time
No non-self incoming citations found for this paper in this database.
Authors
- 1. Chuang Yang (University of Tokyo)
- 2. Renhe Jiang (University of Tokyo)
- 3. Xiaohang Xu (University of Tokyo)
- 4. Chuan Xiao (Osaka University)
- 5. Kaoru Sezaki (University of Tokyo)
BibTeX Citation
@article{yang_vldb25,
title = {{SIMformer: Single-Layer Vanilla Transformer Can Learn Free-Space Trajectory Similarity}},
author = {Yang, Chuang and Jiang, Renhe and Xu, Xiaohang and Xiao, Chuan and Sezaki, Kaoru},
journal = {PVLDB},
series = {{VLDB} '25},
volume = {18},
number = {2},
pages = {390--398},
doi = {10.14778/3705829.3705853},
url = {https://doi.org/10.14778/3705829.3705853},
year = {2025}
}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 7 of 7 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 2,021 | Distributed Trajectory Similarity Search | 2017 | VLDB | 9.1719539e-05 |
| 2,196 | DITA: Distributed In-Memory Trajectory Analytics | 2018 | SIGMOD | 8.8780862e-05 |
| 4,907 | Subtrajectory Clustering: Models and Algorithms | 2018 | PODS | 6.3607956e-05 |
| 5,011 | Fast Large-Scale Trajectory Clustering | 2020 | VLDB | 6.3148258e-05 |
| 5,188 | Fast Subtrajectory Similarity Search in Road Networks under Weighted Edit Distance Constraints | 2020 | VLDB | 6.2370687e-05 |
| 5,743 | VRE: A Versatile, Robust, and Economical Trajectory Data System | 2022 | VLDB | 6.0099017e-05 |
| 6,730 | Trajectory Similarity Measurement: An Efficiency Perspective | 2024 | VLDB | 5.6939772e-05 |
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 2,021 | Distributed Trajectory Similarity Search | 2017 | VLDB |
| 2 | 11,768 | Efficient Non-Learning Similar Subtrajectory Search | 2023 | VLDB |
| 3 | 3,770 | Calibrating Trajectory Data for Similarity-based Analysis | 2013 | SIGMOD |
| 4 | 8,890 | Quantifying Point Contributions: A Lightweight Framework for Efficient and Effective Query-Driven Trajectory Simplification | 2025 | VLDB |
| 5 | 6,890 | On Efficiently Searching Trajectories and Archival Data for Historical Similarities | 2008 | VLDB |
| 6 | 6,031 | Efficient and Effective Similar Subtrajectory Search with Deep Reinforcement Learning | 2020 | VLDB |
| 7 | 10,150 | RED: Effective Trajectory Representation Learning with Comprehensive Information | 2025 | VLDB |
| 8 | 11,255 | SimRN: Trajectory Similarity Learning in Road Networks based on Distributed Deep Reinforcement Learning | 2025 | VLDB |
| 9 | 11,214 | Revisiting CNNs for Trajectory Similarity Learning | 2025 | VLDB |
| 10 | 6,730 | Trajectory Similarity Measurement: An Efficiency Perspective | 2024 | VLDB |