P-BERT: Toward Long Sequence Modeling by Enabling Language Representation With Prefix Sequence Compression, Soft Position Embedding, and Data Augmentation for Patent Relevance Assessment

P-BERT: Toward Long Sequence Modeling by Enabling Language Representation With Prefix Sequence Compression, Soft Position Embedding, and Data Augmentation for Patent Relevance Assessment

Recent works have increasingly adopted pre-trained language models, such as BERT, to model technical semantics for patent relevance assessment. However, existing truncation and divide-merge strategies, used to handle input length constraints, results in feature loss and semantic isolation. This limi...

Full description

Saved in:
Bibliographic Details
Main Authors: Fei Wang, Xingchen Shi, Dongsheng Wang, Yinxia Lou
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Prefix sequence compression
soft position embedding
data augmentation
patent relevance assessment
Online Access:https://ieeexplore.ieee.org/document/10902370/
Tags: Add Tag
No Tags, Be the first to tag this record!
Login for hold and recall information

Internet

https://ieeexplore.ieee.org/document/10902370/

Similar Items