Dongyeop Kang @ UC Berkeley

Dongyeop Kang

I am a postdoctoral scholar at University of California, Berkeley, under Marti A. Hearst. I obtained my Ph.D. in the Language Technologies Institute of the School of Computer Science at Carnegie Mellon University, under Eduard Hovy. I interned at Facebook AI, Allen Institute for AI (AI2), and Microsoft Research. My Ph.D. study has been supported by Allen Institute for AI (AI2) Fellowship, CMU Presidential Fellowship, and ILJU Graduate Fellowship. In the middle of my study, I completed my alternative military service in South Korea at Naver Labs and KAIST Institute. Before joining CMU, I obtained my BS and MS in Computer Science Engineering at KAIST, Korea.

dongyeopk@berkeley.edu

I'm interested in building human-like language generation systems. Natural language generation (NLG) is a key component of many language applications such as dialogue systems, question answering systems, and textual summarization. However, they are yet far behind human-like or human-level generation. This is because a multitude of implicit information is NOT explicitly obvious on the surface. We call the kinds of information as facets, that are reflected in variations of a language, such as external knowledge, intents, interpersonal information, and more. To generate human-like utterances, appropriate modeling of these facets is necessary, and the system needs to be effectively guided by them. Based on Halliday’s Systemic Functional Linguistics (SFL) theory (1978), my Ph.D. thesis focuses on three facet groups; knowledge, structure, and style, and presents effective computational methods for handling each facet in a wide range of generation tasks. During my postdoc, as the intersection of NLG and human-computer interaction, I develop human-machine collaborative language technologies in a mixed-initiative way.

Research Interests:

Computational Linguistics, Human-Computer Interaction, Machine Learning

Neural-symbolic integration for trustworthy and factual generation
Text planning for coherently structured generation
Cross-style language understanding for stylistically appropriate generation
Mixed-initiative interaction for human-in-the-loop generation

News

(Sep 2020) Two papers are accepted to EMNLP 2020.
(Nov 2020) Invited talks at KAIST AI Colloquium and GIST EECS Colloquium.
(Aug 2020) Invited talk about uncertainty calibration in NLP classification tasks at Berkeley AI Research (BAIR) Workshop.
(July 2020) Invited talks at CMU LTI Summer Seminar and POSTECH AI Seminar.
(June 2020) Joined University of California, Berkeley as a postdoctoral scholar, working with Marti A. Hearst.
(May 2020) Passed my Ph.D thesis defense at CMU.

Publication

2020

Plan ahead: Self-Supervised Text Planning for Paragraph Completion Task
Dongyeop Kang, Eduard Hovy
EMNLP 2020 [pdf | arxiv | bib]
INSPIRED: Toward Sociable Recommendation Dialog Systems
Shirley Anugrah Hayati, Dongyeop Kang, Qingxiaoyang Zhu, Weiyan Shi and Zhou Yu
EMNLP 2020 [pdf | arxiv | bib]
Linguistically Informed Language Generation: A Multifaceted Approach
Dongyeop Kang
Committee: Eduard Hovy, Jeffrey Bigham, Alan W Black, Jason Weston, Dan Jurafsky
Ph.D thesis [pdf | bib]
Posterior Calibrated Training on Sentence Classification Tasks
Taehee Jung, Dongyeop Kang, Hua Cheng, Lucas Mentch and Thomas Schaaf
ACL 2020 [pdf | arxiv | code | bib]

2019

xSLUE: A Benchmark and Analysis Platform for Cross-Style Language Understanding and Evaluation
Dongyeop Kang and Eduard Hovy
[pdf | arxiv | data+leaderboard | code | bib]
Earlier Isn't Always Better: Sub-aspect Analysis on Corpus and System Biases in Summarization
Taehee Jung*, Dongyeop Kang*, Lucas Mentch and Eduard Hovy (*equal contribution)
EMNLP 2019 [pdf | arxiv | data+leaderboard | code | bib]
Recommendation as a Communication Game: Self-Supervised Bot-Play for Goal-oriented Dialogue
Dongyeop Kang, Anusha Balakrishnan, Pararth Shah, Paul Crook, Y-Lan Boureau and Jason Weston
EMNLP 2019 [pdf| arxiv | [data] | bib]
(Male, Bachelor) and (Female, Ph.D) have different connotations: Parallelly Annotated Stylistic Language Dataset with Multiple Personas
Dongyeop Kang, Varun Gangal and Eduard Hovy
EMNLP 2019 (Oral) [pdf| arxiv | data+code | slides | bib]
Linguistic Versus Latent Relations for Modeling Coherent Flow in Paragraphs
Dongyeop Kang, Hiroaki Hayashi, Alan W Black, and Eduard Hovy
EMNLP 2019 [pdf | arxiv | code | bib]

2018

Bridging Knowledge Gaps in Neural Entailment via Symbolic Models
Dongyeop Kang, Tushar Khot, Ashish Sabharwal and Peter Clark
EMNLP 2018 [pdf | arxiv | code | bib]
AdvEntuRe: Adversarial Training for Textual Entailment with Knowledge-Guided Examples
Dongyeop Kang, Tushar Khot, Ashish Sabharwal and Eduard Hovy
ACL 2018 [pdf | arxiv | code | bib]
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
NAACL 2018 [pdf | arxiv | data+code | bib]
Actionable email intent modeling with reparametrized RNN
Chu-Cheng Lin, Dongyeop Kang, Michael Gamon, Madian Khabsa, Ahmed Hassan Awadallah, Patrick Pantel
AAAI 2018 [pdf | arxiv | bib]

2017

Detecting and Explaining Causes From Text For a Time Series Event
Dongyeop Kang, Varun Gangal, Ang Lu, Zheng Chen, Eduard Hovy
EMNLP 2017 [pdf | arxiv | dataset | bib]

Before Ph.D. study

News2Images: Automatically Summarizing News Articles into Image-Based Contents via Deep Learning
Jung-Woo Ha, Dongyeop Kang, Hyuna Pyo, Jeonghee Kim
RecSys Workshop 2015 [pdf | bib]
Eventera: Real-time Event Recommendation System from Massive Heterogeneous Online Media
Dongyeop Kang, DongGyun Han, Na Hea Park, Sangtae Kim, U Kang, Soobin Lee
ICDM 2014 (Demo) [pdf | bib | project page | demo]
Data/Feature Distributed Stochastic Coordinate Descent for Logistic Regression
Dongyeop Kang, Woosang Lim, Kijung Shin, Sael Lee, U Kang
CIKM 2014 (Oral) [pdf | bib | appendix | slides]
Hetero-Labeled LDA: A partially supervised topic model with heterogeneous label information
Dongyeop Kang, Youngja Park, Suresh Chari
ECML/PKDD 2014 (Oral) [pdf | bib | slides]
Multidimensional Mining of Large-Scale Search Logs: A Topic-Concept Cube Approach
Dongyeop Kang, Daxin Jiang, Jian Pei, Zhen Liao, Xiaohui Sun, Ho-Jin Choi
WSDM 2011 [pdf | bib | journal version]

Awards

Top Reviewers, NeurIPS 2018 and ACL 2018
Graduate Student Assembly/Provost Conference Funding
Allen Institute for AI (AI2) Fellowship, 2018
CMU Presidential Fellowship, 2017-2018
ILJU Foundation Graduate Scholarship, 2011-2016
SMBA Startups Investment Prize, 2013

Experience

University of California, Berkeley, postdoctoral scholar (PI: Marti A. Hearst), 2020-Present, CA, USA
CMU, research assistant (advisor: Eduard Hovy), 2015-2020, PA, USA
Facebook AI, research intern (collaborators: Anusha Balakrishnan, Pararth Shah, Rajen Subba, Paul Crook, Y-Lan Boureau, Jason Weston), 2018, CA, USA
Allen Institute for Artificial Intelligence (AI2), research intern (collaborators: Tushar Khot, Ashish Sabharwal, Peter Clark), 2017, WA, USA
Microsoft Research, research intern (collaborators: Michael Gamon, Patrick Pantel), 2016, WA, USA
Eventrader, founder, 2015, Seoul, Korea
Naver Labs, researcher, 2014-2015, Seoul, Korea
Oracle Labs, research intern (mentor: Sungpack Hong), 2014, CA, USA.
IBM TJ Watson Research, research intern (mentor: Youngja Park), 2013, NY, USA.
KAIST Institute, researcher, 2012-2014, Seoul, Korea
CMU / PanOptus, research assistant (advisor: Eric Xing), 2011-2012, PA, USA
HKUST, research assistant (advisor: Sung Kim), 2011, Hong Kong.
Microsoft Research, research intern (mentor: Ed Nightingale), 2010, WA, USA.
Microsoft Research Asia, research intern (mentor: Daxing Jiang, Hang Li), 2009-2010, Beijing, China.
ETRI, intern, 2008, Daejeon, Korea.
DML/CDSN/KECI Labs at KAIST, research assistant, 2007-2009, Daejeon, Korea.

Services

Program Committee / Reviewer of ICLR21
Program Committee / Reviewer of ICLR20, ACL20 (Generation track), NeurIPS20, EMNLP20 (Generation track)
Program Committee / Reviewer of ICLR19, ICML19, NAACL19 (Style track), ACL19 (Machine Learning / Generation / Question Answering / Sentence-level Semantics / Applications track), EMNLP19 (Lexical Semantics track), NeurIPS19, W-NUT19, Scientometrics19
Program Committee / Reviewer of NeurIPS18 (top-30% reviewer), EMNLP18 (Discourse track), ACL18 (Machine Learning track) (top-reviewer), MRQA18

Talks and Teaching

Invited talk on "Linguistically-informed Language Generation" at CMU LTI Seminar, 2020 Summer
Invited talk on "Toward Human-like Language Generation" at POSTECH AI Seminar, 2020 Summer
Lecture on "Relational Semantics on Word Embeddings" at Computational Semantics for NLP (advisor: Eduard Hovy, Teruko Mitamura), 2019 Spring
CMU, teaching assistant, Computational Semantics for NLP (advisor: Eduard Hovy, Teruko Mitamura), 2018 Spring

Lecture on "Distributional Compositionality and Logics"

CMU, teaching assistant, Machine Translation and Sequence to Sequence Models (advisor: Graham Neubig), 2017 Spring

Last updated in June 2020