On the Limitations and Capabilities of Position Embeddings for Length Generalization
IEEE Transactions on Pattern Analysis and Machine Intelligence · 2026-09 · 48 (9) · 10844-10855
DOI: 10.1109/tpami.2026.3688702 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Pattern Analysis and Machine Intelligence, 2026-09. Verfasst von Yang Chen und 2 weiteren Autor:innen. Die Arbeit nennt 29 Literaturangaben. Gefördert von: NSF China, Beijing Natural Science Foundation, Beijing Major Science and Technology Project.
Abstract (Originalsprache)
Kein Abstract verfügbar.
Förderung
- NSF China
- Beijing Natural Science Foundation
- Beijing Major Science and Technology Project
Zitieren
APA
Chen, Y., Liang, Y., & Lin, Z. (2026). On the Limitations and Capabilities of Position Embeddings for Length Generalization. IEEE Transactions on Pattern Analysis and Machine Intelligence, 48(9), 10844-10855. https://doi.org/10.1109/tpami.2026.3688702
Vancouver
Chen Y, Liang Y, Lin Z. On the Limitations and Capabilities of Position Embeddings for Length Generalization. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2026;48(9):10844-10855. doi:10.1109/tpami.2026.3688702
BibTeX
@article{Chen2026,
author = {Chen, Yang and Liang, Yitao and Lin, Zhouchen},
title = {On the Limitations and Capabilities of Position Embeddings for Length Generalization},
journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
year = {2026},
volume = {48},
number = {9},
pages = {10844--10855},
doi = {10.1109/tpami.2026.3688702}
}
EUVIMED