Mind the Links: Cross-Layer Attention for Link Prediction in Multiplex Networks DOI: XXXXXXX.XXXXXXX DOI: 10.1145/XXXXXXX.XXXXXXX Conference: ; ; ISBN: 978-1-4503-XXXX-X/26/03 ISBN: 978-1-4503-XXXX-X/2018/06
Devesh Sharma, Aditya Kishore, Ayush Garg, Debajyoti Mazumder, Debasis Mohapatra, Jasabanta Patro · 2025
Read Mind the Links: Cross-Layer Attention for Link Prediction in Multiplex Networks DOI: XXXXXXX.XXXXXXX DOI: 10.1145/XXXXXXX.XXXXXXX Conference: ; ; ISBN: 978-1-4503-XXXX-X/26/03 ISBN: 978-1-4503-XXXX-X/2018/06 by Devesh Sharma, Aditya Kishore, Ayush Garg, Debajyoti Mazumder, Debasis Mohapatra, Jasabanta Patro (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-23409.txt.
Description
Networks with many relation types, like work plus family plus school ties, need link forecasts that read all layers at once. Devesh Sharma and 5 co-authors frame that task as multi-view edge classification. Each node pair yields one edge view per layer. Cross-layer self-attention fuses those views for the target layer. Two models fill the frame: a light transformer on static embeddings, and a graph-attention hybrid.
A union candidate pool and leak-free protocols keep tests fair and scalable. Six public multiplex data sets frame the trials. Both models beat strong baselines on macro F1 by consistent margins. The frame accepts precomputed embeddings or graph encoders alike.
Catalog record
| CREATOR | Devesh Sharma, Aditya Kishore, Ayush Garg, Debajyoti Mazumder, Debasis Mohapatra, Jasabanta Patro |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 19 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2509-23409