Richard Yuanzhe Pang

New York University
yzpang.me
Publications18
h-index8
Citations268
Highly Influential Citations24
Author pages are created from data sourced from our academic publisher partnerships and public sources.

Recommended Authors

  • Samuel R. Bowman
  • 95 Publications • 15,163 Citations
  • Kyunghyun Cho
  • 323 Publications • 78,754 Citations
  • Luke Zettlemoyer
  • 232 Publications • 40,171 Citations
  • Kevin Gimpel
  • 126 Publications • 12,259 Citations
  • Publications
  • Influence
Intermediate-Task Transfer Learning with Pretrained Language Models: When and Why Does It Work?
TLDR
It is observed that intermediate tasks requiring high-level inference and reasoning abilities tend to work best and that target task performance is strongly correlated with higher-level abilities such as coreference resolution, but it is failed to observe more granular correlations between probing and target taskperformance.
View on ACL
Intermediate-Task Transfer Learning with Pretrained Models for Natural Language Understanding: When and Why Does It Work?
TLDR
It is observed that intermediate tasks requiring high-level inference and reasoning abilities tend to work best and that target task performance is strongly correlated with higher-level abilities such as coreference resolution, but it is failed to observe more granular correlations between probing and target taskperformance.
View PDF on arXiv
Text Generation by Learning from Demonstrations
View Paper
Consistency of a Recurrent Language Model with Respect to Incomplete Decoding
TLDR
It is proved that commonly used incomplete decoding algorithms - greedy search, beam search, top-k sampling, and nucleus sampling - are inconsistent, despite the fact that recurrent language models are trained to produce sequences of finite length.
View on ACL
ENGINE: Energy-Based Inference Networks for Non-Autoregressive Machine Translation
TLDR
This work views their non-autoregressive translation system as an inference network trained to minimize the autoregressive teacher energy, which achieves state-of-the-art non-Autoregressive results on the IWSLT 2014 DE-EN and WMT 2016 RO-EN datasets, approaching the performance of autore progressive models.
View on ACL
Unsupervised Evaluation Metrics and Learning Criteria for Non-Parallel Textual Transfer
TLDR
This work considers the problem of automatically generating textual paraphrases with modified attributes or properties, focusing on the setting without parallel data, and proposes additional metrics based on semantic preservation and fluency as well as a way to combine them into a single overall score.
View on ACL
Towards Actual (Not Operational) Textual Style Transfer Auto-Evaluation
TLDR
The dangerous current state of style transfer auto-evaluation research is elucidated and ways to aggregate the three metrics into one evaluator are proposed.
View on ACL
AgreeSum: Agreement-Oriented Multi-Document Summarization
TLDR
This work creates a dataset for AgreeSum, and provides annotations on article-summary entailment relations for a subset of the clusters in the dataset, and hopes that these article- summary entailment annotations contribute to the community’s effort in improving abstractive summarization faithfulness.
View on ACL
QuALITY: Question Answering with Long Input Texts, Yes!
TLDR
QuALITY, a multiple-choice QA dataset with context passages in English that have an average length of about 5,000 tokens, much longer than typical current models can process, is introduced to enable building and testing models on long-document comprehension.
View PDF on arXiv
Comparing Test Sets with Item Response Theory
TLDR
Quoref, HellaSwag, and MC-TACO are best suited for distinguishing among state-of-the-art models, while SNLI, MNLI, and CommitmentBank seem to be saturated for current strong models.
View on ACL
...
...