Projects: pypi: gpl

https://packages.ecosyste.ms/registries/pypi.org/packages/gpl

GPL is an unsupervised domain adaptation method for training dense retrievers. It is based on query generation and pseudo labeling with powerful cross-encoders. To train a domain-adapted model, it needs only the unlabeled target corpus and can achieve significant improvement over zero-shot models.
14 versions
Latest release: over 3 years ago
309 downloads last month

Enhanced Analysis
Repository Activity:
Repository Owner: Ubiquitous Knowledge Processing Lab (organization) Academic
Science Score: 100/100
Starting Score: 100 points
Bonuses:
  • +20 Academic repository owner
    Repository owned by academic institution
  • +15 Institutional repository owner
    Repository owned by research institution
Penalties:
  • -10 PyPI ecosystem
    General-purpose ecosystem
  • -15 No science keywords
    No scientific terms found in keywords/classifiers
  • -10 No science terms in description
    No scientific terms found in description

Very Likely Science (100)

Papers Mentioning gpl 1

10.18632/aging.103350
A novel molecular-clinicopathologic nomogram to improve prognosis prediction of hepatocellular carcinoma
Cited by: 3
Author(s): Zhongjing Zhang, Wanqing Weng, Weiguo Huang, Boda Wu, Yi Zhou, Jie Zhang, Tuo Deng, Yan Wen, Jiecheng Zhang, Jian‐Yang Ao, Qiyu Zhang, Ke‐Qing Shi
Software Mentions: 6
Published: almost 6 years ago