D
Dongyeop Kang
Researcher at University of California, Berkeley
Publications - 74
Citations - 832
Dongyeop Kang is an academic researcher from University of California, Berkeley. The author has contributed to research in topics: Computer science & Task (project management). The author has an hindex of 12, co-authored 48 publications receiving 464 citations. Previous affiliations of Dongyeop Kang include Carnegie Mellon University & University of Pittsburgh.
Papers
More filters
Proceedings ArticleDOI
A Dataset of Peer Reviews (PeerRead): Collection, Insights and NLP Applications
Dongyeop Kang,Waleed Ammar,Bhavana Dalvi,Madeleine van Zuylen,Sebastian Kohlmeier,Eduard Hovy,Roy Schwartz,Roy Schwartz +7 more
TL;DR: The first public dataset of scientific peer reviews available for research purposes (PeerRead v1) is presented and it is shown that simple models can predict whether a paper is accepted with up to 21% error reduction compared to the majority baseline.
Proceedings ArticleDOI
AdvEntuRe: Adversarial Training for Textual Entailment with Knowledge-Guided Examples
TL;DR: The authors propose knowledge-guided adversarial example generators for incorporating large lexical resources in entailment models via only a handful of rule templates, and make the entailment model more robust by iteratively adjusting to the discriminator's weaknesses.
Posted Content
Recommendation as a Communication Game: Self-Supervised Bot-Play for Goal-oriented Dialogue
TL;DR: This work collects a goal-driven recommendation dialogue dataset (GoRecDial), which consists of 9,125 dialogue games and 81,260 conversation turns between pairs of human workers recommending movies to each other, and uses the dataset to develop an end-to-end dialogue system that can simultaneously converse and recommend.
Posted Content
A Dataset of Peer Reviews (PeerRead): Collection, Insights and NLP Applications
Dongyeop Kang,Waleed Ammar,Bhavana Dalvi,Madeleine van Zuylen,Sebastian Kohlmeier,Eduard Hovy,Roy Schwartz,Roy Schwartz +7 more
TL;DR: The PeerRead v1 dataset as discussed by the authors contains 14.7k paper drafts and corresponding accept/reject decisions in top-tier venues including ACL, NIPS and ICLR.
Proceedings ArticleDOI
GenAug: Data Augmentation for Finetuning Text Generators
TL;DR: This paper proposes and evaluates various augmentation methods, including some that incorporate external knowledge, for finetuning GPT-2 on a subset of Yelp Reviews, and examines the relationship between the amount of augmentation and the quality of the generated text.