Assistant Professor
Computer Science & Engineering
University of Minnesota, Twin Cities
259 Shepherd
Minnesota NLP group (under construction)
Minnesota NLP seminar
MnRI / MCAL / DSI / TextGroup / LIL
dongyeop@umn.edu
Office hour (by appointment)
University of Minnesota, Twin Cities
259 Shepherd
Minnesota NLP seminar
MnRI / MCAL / DSI / TextGroup / LIL
Office hour (by appointment)
I'm interested in developing human-centered language technologies.
I obtained a Ph.D. on natural language generation at Carnegie Mellon University under Eduard Hovy.
I did my postdoc at UC Berkeley, under Marti A. Hearst, for extending NLG/NLP systems to be more interactive.
Our Minnesota NLP group's current research is focused on building human-centric NLP systems by learning from human perception, interaction, and disagreements:
- Scaffold NLP systems with human cognition by learning from annotated explanations, eye-tracking, and implicit feedback
- Improve productivity and creativity of scientific reading/writing with interactive and collaborative NLP systems
- Build inclusive and individualized NLP systems learning from disagreements among humans
News (read more)
- (Jan 2023) Very excited to receive a MnRI seed grant (PI: Karthik Desingh, Co-PI: Dongyeop Kang)
, working on Robotics+NLP. - (Nov 2022) Our second In2Writing workshop will be held at CHI 2023 in Hamburg, Germany.
- (Oct 2022) Very excited to receive a generous gift from Grammarly
.
- (Oct 2022) Minnesota NLP seminar is back in Fall 2022!
- (Jun 2022) Very honored to receive the 3M Non-tenured Faculty Award (NTFA)
. - (Jun 2022) Very honored to receive the Sony Research Faculty Innovation Award
.
- (May 2022) Received the best paper award on human-in-the-loop interative text revision from In2Writing workshop at ACL 2022
- (May 2022) At ACL 2022, our group presents two papers ( iterative text revision and human-in-the-loop interative text revision) and co-organizes Intelligent and Interactive Writing Assistants (In2Writing) workshop
- (Feb 2022) Four Ph.D. students are co-organizing Minnesota NLP seminar for Spring 2022
- (Nov 2021) I am co-organizing "CtrlGen: Controllable Generative Modeling in Language and Vision" workshop at NeurIPS 2021.
- Three papers (1 main, 1 finding, and 1 workshop) from Minnesota NLP will be presented at EMNLP 2021.
- (Apr 2021) I will be joining Minnesota CSE as an assistant professor in 2021 Fall.
- (Jun 2020) Joined UC Berkeley as a postdoctoral scholar, working with Marti A. Hearst.
- (May 2020) Passed my Ph.D thesis defense. Check out my thesis!
Talks (read more)
Talk at Microsoft Research, May 2023
Talk at Google People+AI Research (PAIR), Feb 2023
Talk at Hyundai AI, Dec 2022
Talk at NSF ROSE-HUB, Nov 2022
Talk at 3M NTFA Symposium, Sep 2022
Talk at Yonsei University, Aug 2022
Talk at KAIST CS, Aug 2022
Talk at LG AI Research, Aug 2022
Talk at Samsung Research, Aug 2022
Talk at Grammarly, Jul 2022
Talk at Thomson Reuters Labs, Jul 2022
Talk at USC ISI AI Seminar, Mar 2022
Talk and panel discussion at CMU LTI Seminar, Feb 2022
Talk at GaTech NLP Seminar, Dec 2021
Talk at Minnesota Robotic Institute Colloquium, Nov 2021
Talk at KAIST SE Colloquium, Nov 2021
Talk at Naver Labs Europe, Nov 2021
Talk at Grammarly AI, Aug 2021
Talk at SNU Summer AI school, Aug 2021
Talk at University of Wisconsin - Madison, CS Department, March 2021
Talk at University of Rochester, CS Department, March 2021
Talk at University of Minnesota - Twin Cities, CSE Colloquium, March 2021
Talk at KAIST, AI Department, Feb 2021
Talk at KAIST AI Colloquium, Nov 2020
Talk at GIST EECS Colloquium, Nov 2020
Talk at Berkeley AI Research (BAIR) Workshop, July 2020
Talk at CMU LTI Seminar, July 2020
Talk at POSTECH AI Seminar, July 2020
Current students and visitors
PhD
PhD
w/ Jaideep Srivastava
PhD
w/ Maria Gini
visiting PhD @KAIST
advisor: Jinwoo Shin
PhD
w/ Maria Gini, CSE fellow
PhD
3M fellow
PhD
PhD
CSE fellow
MS
Undergraduate
Undergraduate
You can find the full list of students, alumni, and collaborators in Minnesota NLP group.
Our research is supported by
.
If you like to join us, please read this page and fill out this form.
NOTE: If you are already being advised by someone, please talk to your advisor first and ask them to contact me.
Publication ( Selected | By Date | By Topic | Google Scholar )
2023
-
Decoding the End-to-end Writing Trajectory in Scholarly Manuscripts
2022
-
Improving Iterative Text Revision by Learning Where to Edit from other Revision Tasks
-
User or Labor: An Interaction Framework for Human-Machine Relationships in NLP
-
Read, Revise, Repeat: A System Demonstration for Human-in-the-loop Iterative Text RevisionBest paper award
2021
-
Does BERT Learn as Humans Perceive? Understanding Linguistic Styles through Lexica
-
Modeling Mathematical Notation Semantics in Academic Papers
-
Visualizing Cross-Lingual Discourse Relations in Multilingual TED Corpora
-
Understanding Out-of-distribution: A Perspective of Data Dynamics
-
Style is NOT a single variable: Case Studies for Cross-Stylistic Language Understanding
2020
-
GenAug: Data Augmentation for Finetuning Text Generators
-
Document-Level Definition Detection in Scholarly Documents: Existing Models, Error Analyses, and Future Directions
-
Plan ahead: Self-Supervised Text Planning for Paragraph Completion Task
-
INSPIRED: Toward Sociable Recommendation Dialog Systems
-
Linguistically Informed Language Generation: A Multifaceted Approach
2019
-
Earlier Isn't Always Better: Sub-aspect Analysis on Corpus and System Biases in Summarization
-
Recommendation as a Communication Game: Self-Supervised Bot-Play for Goal-oriented Dialogue
-
(Male, Bachelor) and (Female, Ph.D) have different connotations: Parallelly Annotated Stylistic Language Dataset with Multiple Personas
-
Linguistic Versus Latent Relations for Modeling Coherent Flow in Paragraphs
2018
-
Bridging Knowledge Gaps in Neural Entailment via Symbolic Models
-
A Dataset of Peer Reviews (PeerReaD): Collection, Insights and NLP Applications
2017
-
Detecting and Explaining Causes From Text For a Time Series Event
2011-2015
-
News2Images: Automatically Summarizing News Articles into Image-Based Contents via Deep Learning
-
Eventera: Real-time Event Recommendation System from Massive Heterogeneous Online Media
-
Hetero-Labeled LDA: A partially supervised topic model with heterogeneous label information
-
Multidimensional Mining of Large-Scale Search Logs: A Topic-Concept Cube Approach
-
Read, Revise, Repeat: A System Demonstration for Human-in-the-loop Iterative Text RevisionBest paper award
-
Does BERT Learn as Humans Perceive? Understanding Linguistic Styles through Lexica
-
Style is NOT a single variable: Case Studies for Cross-Stylistic Language Understanding
-
Linguistically Informed Language Generation: A Multifaceted Approach
-
Recommendation as a Communication Game: Self-Supervised Bot-Play for Goal-oriented Dialogue
-
A Dataset of Peer Reviews (PeerReaD): Collection, Insights and NLP Applications
Discourse modeling and planning
-
Improving Iterative Text Revision by Learning Where to Edit from other Revision Tasks
-
Plan ahead: Self-Supervised Text Planning for Paragraph Completion Task
-
INSPIRED: Toward Sociable Recommendation Dialog Systems
-
Recommendation as a Communication Game: Self-Supervised Bot-Play for Goal-oriented Dialogue
-
Linguistic Versus Latent Relations for Modeling Coherent Flow in Paragraphs
-
Detecting and Explaining Causes From Text For a Time Series Event
Linguistic styles and pragmatics
-
Does BERT Learn as Humans Perceive? Understanding Linguistic Styles through Lexica
-
Style is NOT a single variable: Case Studies for Cross-Stylistic Language Understanding
-
(Male, Bachelor) and (Female, Ph.D) have different connotations: Parallelly Annotated Stylistic Language Dataset with Multiple Personas
Ethics and bias
-
Earlier Isn't Always Better: Sub-aspect Analysis on Corpus and System Biases in Summarization
Neural-symbolic representation and reasoning
-
GenAug: Data Augmentation for Finetuning Text Generators
-
Bridging Knowledge Gaps in Neural Entailment via Symbolic Models
Interactive and collaborative NLP systems
-
Read, Revise, Repeat: A System Demonstration for Human-in-the-loop Iterative Text RevisionBest paper award
-
Modeling Mathematical Notation Semantics in Academic Papers
-
Document-Level Definition Detection in Scholarly Documents: Existing Models, Error Analyses, and Future Directions
-
A Dataset of Peer Reviews (PeerReaD): Collection, Insights and NLP Applications
Data-centric NLP
Others
-
News2Images: Automatically Summarizing News Articles into Image-Based Contents via Deep Learning
-
Eventera: Real-time Event Recommendation System from Massive Heterogeneous Online Media
-
Hetero-Labeled LDA: A partially supervised topic model with heterogeneous label information
-
Multidimensional Mining of Large-Scale Search Logs: A Topic-Concept Cube Approach