Peilin Yu

CIT 351

115 Waterman St

Providence, RI 02905

I am a Ph.D. Candidate in Computer Science advised by Prof. Stephen Bach.

My main focus is on tackling weak supervision problems using approaches that involve working with limited, indirect, and/or noisy supervision while minimizing human effort. Additionally, I investigate the benefits of leveraging foundation models (GPTs, T5s, CLIP etc.) to enhance weak supervision workflow.

I am a proud 👐🏟🧀 Badger and I got my BS in CS and Math there in 2019.

Email : [first_name]_[last_name] [at] brown [dot] edu

news

Feb 27, 2023	We released Alfred, a toolkit for integrating Large Pretrained Models into Weak Supervision

selected publications

AISTATS

Learning from Multiple Noisy Partial Labelers

Yu, Peilin, Ding, Tiffany, and Bach, Stephen H.

In Artificial Intelligence and Statistics (AISTATS) 2022

Bib PDF Code

@inproceedings{yu2022nplm,
  abbr = {AISTATS},
  title = {Learning from Multiple Noisy Partial Labelers},
  author = {Yu, Peilin and Ding, Tiffany and Bach, Stephen H.},
  booktitle = {Artificial Intelligence and Statistics (AISTATS)},
  year = {2022},
  bibtex_show = {true},
  selected = {true},
  pdf = {2106.04530.pdf},
  code = {https://github.com/batsresearch/nplm}
}

ICLR

Learning to Compose Soft Prompts for Compositional Zero-Shot Learning

Yu, Peilin*, Nayak, Nihal V.*, and Bach, Stephen H. (* Co-first Author)

International Conference on Learning Representations (ICLR) 2023

PDF Code

ACL

DIAG-NRE: A Neural Pattern Diagnosis Framework for Distantly Supervised Neural Relation Extraction

Zheng, Shun, Han, Xu, Lin, Yankai, Yu, Peilin, Chen, Lu, Huang, Ling, Liu, Zhiyuan, and Xu, Wei

In ACL 2019

Bib PDF Code

@inproceedings{zheng2019diagnre,
  abbr = {ACL},
  title = {DIAG-NRE: A Neural Pattern Diagnosis Framework for Distantly Supervised Neural Relation Extraction},
  author = {Zheng, Shun and Han, Xu and Lin, Yankai and Yu, Peilin and Chen, Lu and Huang, Ling and Liu, Zhiyuan and Xu, Wei},
  booktitle = {ACL},
  year = {2019},
  bibtex_show = {true},
  selected = {true},
  pdf = {1811.02166.pdf},
  code = {https://github.com/thunlp/DIAG-NRE}
}