Crossmark

Transformer++: a long sequence modeling method based on direction-aware dual attention and multi-head sampling

Published Online: 2025-11-19

Published Print: 2025-11

Authors

Wang, Ruiqin https://orcid.org/0000-0003-3034-5467
Ji, Qishun

Sheng, Zhenzhen

Qi, Yang
Funding

Funding for this research was provided by:

National Natural Science Foundation of China (62277016)
License Information

Text and Data Mining valid from 2025-11-01

Version of Record valid from 2025-11-01
More Information

Article History

Received: 29 December 2024

Accepted: 17 October 2025

First Online: 19 November 2025

Declarations

:

: The authors declare no conflict of interest.

Document is current