Papers by Dongyeop Kang

(Male, Bachelor) and (Female, Ph.D) have different connotations: Parallelly Annotated Stylistic Language Dataset with Multiple Personas (D19-1)

Challenge:	a novel approach to paragraph planning involves a high-level control of different levels of relations between sentences . a proposed model with both forms of relations outperforms baselines in partially conditioned paragraph generation task .
Approach:	They propose two models that integrate human-created and latent relations into document-level language models . they focus on paragraph-level plan between sentences to produce coherent text .
Outcome:	The proposed models outperform baselines in partially conditioned paragraph generation task.

Copied to clipboard

Dongyeop Kang, Varun Gangal, Eduard Hovy

Challenge:	Despite recent attempts on computational modeling of the variation, the lack of parallel corpora of style language makes it difficult to systematically control the stylistic change and evaluate such models.
Approach:	They propose to use a parallel and annotated stylistic language dataset to test the effectiveness of style transfer models.
Outcome:	The proposed model outperforms the unsupervised models using nonparallel corpus.

CoEdIT: Text Editing by Task-Specific Instruction Tuning (2023.findings-emnlp)

Copied to clipboard

Vipul Raheja, Dhruv Kumar, Ryan Koo, Dongyeop Kang

Abstain-R1: Calibrated Abstention and Post-Refusal Clarification via Verifiable RL (2026.findings-acl)

Challenge:	We present a large language model for writing assistance that is fine-tuned on task-specific instructions.
Approach:	They propose a large language model that is fine-tuned on task-specific instructions and outputs the edited text.
Outcome:	The proposed model performs better than other state-of-the-art models on various editing benchmarks while being 60x smaller.

Copied to clipboard

Haotian Zhai, Jingcheng Liang, Dongyeop Kang

Challenge:	Existing abstention methods produce generic refusals or encourage follow-up clarifications without verifying whether they identify the key missing information.
Approach:	They propose a clarification-aware RLVR reward that rewards correct answers on unanswerable queries while optimizing explicit abstention and semantically aligned post-refusal clarification on unannounced queries.
Outcome:	The proposed model improves abstention and clarification on unanswerable queries while maintaining strong performance on answerable queries.

How LLMs Comprehend Temporal Meaning in Narratives: A Case Study in Cognitive Evaluation of LLMs (2025.acl-long)

Copied to clipboard

Karin De Langis, Jong Inn Park, Andreas Schramm, Bin Hu, Khanh Chi Le, Dongyeop Kang

Challenge:	Large language models exhibit increasingly sophisticated linguistic capabilities, yet the extent to which these models reflect human-like cognition versus advanced pattern recognition remains an open question.
Approach:	They conduct a series of targeted experiments to assess whether LLMs construct semantic representations and pragmatic inferences in a human-like manner.
Outcome:	The proposed framework can be used to assess the cognitive and linguistic capabilities of large language models (LLMs).

Style is NOT a single variable: Case Studies for Cross-Stylistic Language Understanding (2021.acl-long)

Copied to clipboard

Dongyeop Kang, Eduard Hovy

Information Extraction & Text Mining

Challenge:	a benchmark corpus of text in 15 different styles is used to study stylistic language . a similar benchmark is used for cross-style language understanding .
Approach:	They propose a benchmark corpus that combines existing datasets and collects a new one for cross-style language understanding.
Outcome:	The proposed benchmark corpus contains 15 different styles under four theoretical groupings: figurative, personal, affective, and interpersonal groups.

Recommendation as a Communication Game: Self-Supervised Bot-Play for Goal-oriented Dialogue (D19-1)

Copied to clipboard

Dongyeop Kang, Anusha Balakrishnan, Pararth Shah, Paul Crook, Y-Lan Boureau, Jason Weston

Dialogue Systems & Conversational Agents

Challenge:	Traditional recommendation systems produce static rather than interactive recommendations invariant to a user’s specific requests, clarifications, or current mood.
Approach:	They use a goal-driven recommendation dialogue dataset to develop an end-to-end dialogue system that can simultaneously converse and recommend.
Outcome:	The proposed system can converse and recommend movies to humans without considering the task goal itself.

Cluster-Guided Label Generation in Extreme Multi-Label Classification (2023.eacl-main)

Copied to clipboard

Taehee Jung, Joo-kyung Kim, Sungjin Lee, Dongyeop Kang

Information Extraction & Text Mining

Challenge:	Existing classification-based models are poorly per-form for tail labels and ignore semantic relations among labels.
Approach:	They propose to guide label generation using label cluster information to hierarchically generate lower-level labels.
Outcome:	The proposed model outperforms classification and generation baselines on tail labels and improves in four popular XMC benchmarks.

Improving Iterative Text Revision by Learning Where to Edit from Other Revision Tasks (2022.emnlp-main)

Copied to clipboard

Zae Myung Kim, Wanyu Du, Vipul Raheja, Dhruv Kumar, Dongyeop Kang

Syntactic Text Processing

Challenge:	Iterative text revision improves text quality by fixing grammatical errors, rephrasing for better readability or contextual appropriateness.
Approach:	They propose to build an end-to-end text revision system that can iteratively generate helpful edits by explicitly detecting editable spans with their corresponding edit intents.
Outcome:	The proposed system outperforms baselines on other text revision tasks and human evaluations.

When Thoughts Meet Facts: Reusable Reasoning for Long-Context LMs (2026.findings-acl)

Copied to clipboard

Soyeong Jeong, Taehee Jung, Sung Ju Hwang, Joo-Kyung Kim, Dongyeop Kang

Benchmarking Cognitive Biases in Large Language Models as Evaluators (2024.findings-acl)

Challenge:	Recent Long-Context Language Models (LCLMs) do not capture how evidence should be connected . a new framework that integrates thought templates into LCLM frameworks is proving useful .
Approach:	They propose a framework that iteratively refines reusable reasoning patterns derived from prior problem solving to improve their templates.
Outcome:	The proposed framework outperforms baselines on knowledge-intensive multi-hop reasoning benchmarks and practical scenarios without retrieval.

Copied to clipboard

Ryan Koo, Minhwa Lee, Vipul Raheja, Jong Inn Park, Zae Myung Kim, Dongyeop Kang

Bridging Knowledge Gaps in Neural Entailment via Symbolic Models (D18-1)

Challenge:	Large Language Models (LLMs) have been shown to be effective as automatic evaluators with simple prompting and in-context learning.
Approach:	They assemble 16 Large Language Models and evaluate their outputs by preference ranking . they introduce a cognitive bias benchmark to measure six different cognitive biases in LLM evaluation outputs.
Outcome:	The proposed model is biased on the CoBBLer benchmark, indicating that machine preferences are misaligned with humans.

Copied to clipboard

Dongyeop Kang, Tushar Khot, Ashish Sabharwal, Peter Clark

Reasoning

Challenge:	Textual entailment models focus on lexical gaps but rarely on knowledge gaps.
Approach:	They propose a fact-level decomposition of the hypothesis and a knowledge lookup module to fill knowledge gaps in Science Entailment task.
Outcome:	The proposed model outperforms the base model on the SciTail dataset by 3% and 5% on the textual premise and the structured knowledge base.

BBScoreV2: Learning Time-Evolution and Latent Alignment from Stochastic Representation (2025.emnlp-main)

Copied to clipboard

Tianhao Zhang, Zhecheng Sheng, Zhexiao Lin, Chen Jiang, Dongyeop Kang

Representation Learning

Challenge:	Autoregressive generative models are gaining traction in language tasks such as text generation and machine translation.
Approach:	They propose a likelihood-based evaluation metric that fits transformer-based model embeddings into a stochastic process and propose it as a probability-based metric.
Outcome:	The proposed model embeddings induce a "clustered-to-temporal ordered" mapping of language model representations in high-dimensional space, and this structure enhances performance on tasks such as temporal consistency evaluation and AI-generated content detection.

Does BERT Learn as Humans Perceive? Understanding Linguistic Styles through Lexica (2021.emnlp-main)

Copied to clipboard

Shirley Anugrah Hayati, Dongyeop Kang, Lyle Ungar

Rethinking Annotation: Can Language Learners Contribute? (2023.acl-long)

Challenge:	Using pre-trained models, people use different styles to express their interpersonal goal and attitude in their communication.
Approach:	They use a dataset to collect lexicon usages across styles using two lenses: human perception and machine word importance.
Outcome:	The proposed model can predict human perception and machine word importance based on a popular style classifier like BERT . human- and machine-identified words share significant overlap for some styles .

Copied to clipboard

Haneul Yoo, Rifki Afina Putri, Changyoon Lee, Youngin Lee, So-Yeon Ahn, Dongyeop Kang, Alice Oh

Challenge:	Researchers have traditionally recruited native speakers to provide annotations for benchmark datasets, but there are languages for which recruiting native speakers is difficult.
Approach:	They recruit 36 language learners and provide two types of additional resources and perform mini-tests to measure their language proficiency.
Outcome:	The proposed method improves learners' language proficiency in terms of vocabulary and grammar.

Scaling Unverifiable Rewards: A Case Study on Visual Insights (2026.findings-acl)

Copied to clipboard

Shuyu Gan, James Mooney, Pan Hao, Renxiang Wang, Mingyi Hong, Qianwen Wang, Dongyeop Kang

How Far Can We Extract Diverse Perspectives from Large Language Models? (2024.emnlp-main)

Challenge:	Existing methods to scale complex, open-ended tasks with unverifiable rewards are not scalable to multi-stage pipelines.
Approach:	They propose a process-based refinement framework that scales inference across stages of a multi-agent pipeline, instead of refining a single output over time.
Outcome:	The proposed framework scales inference across stages of a multi-agent pipeline, instead of refining a single output over time as in prior work.

Copied to clipboard

Shirley Hayati, Minhwa Lee, Dheeraj Rajagopal, Dongyeop Kang

Earlier Isn’t Always Better: Sub-aspect Analysis on Corpus and System Biases in Summarization (D19-1)

Challenge:	Recent advances of large language models have gained much interest from researchers to exploit their capability of creative generation for data augmentation with less cost and higher diversity.
Approach:	They propose a criteria-based prompting technique to extract maximum diversity from LLMs.
Outcome:	The proposed method extracts diverse opinions from large language models iteratively.

Copied to clipboard

Taehee Jung, Dongyeop Kang, Lucas Mentch, Eduard Hovy

Summarization

Challenge:	despite recent advances in neural summarization systems, the underlying logic behind the improvements remains unexplored.
Approach:	They define three sub-aspects of summarization: position, importance, diversity . position exhibits substantial bias in news articles, but not with academic papers .
Outcome:	evaluators found that position bias is not present in academic papers and meeting minutes . elucidation provides useful lessons on analyzing summarization datasets .

A Dataset of Peer Reviews (PeerRead): Collection, Insights and NLP Applications (N18-1)

Copied to clipboard

Dongyeop Kang, Waleed Ammar, Bhavana Dalvi, Madeleine van Zuylen, Sebastian Kohlmeier, Eduard Hovy, Roy Schwartz

Challenge:	a dataset of 14.7K paper drafts and accept/reject decisions in top-tier venues including ACL, NIPS and ICLR is presented to study peer reviews.
Approach:	They propose to use the dataset to collect peer reviews from top-tier venues including ACL, NIPS and ICLR and to use it to create a dataset of peer reviews for research purposes.
Outcome:	The proposed dataset includes 14.7K paper drafts and accept/reject decisions in top-tier venues including ACL, NIPS and ICLR.

ScholaWrite: A Dataset of End-to-End Scholarly Writing (2026.acl-long)

Copied to clipboard

Khanh Chi Le, Linghe Wang, Minhwa Lee, Ross Volkov, Luan Tuyen Chau, Dongyeop Kang

Challenge:	SCHOLAWRITE traces the multi-month journey from initial drafts to final manuscripts . authors demonstrate the value of capturing scientists’ cognitive writing process .
Approach:	They present a dataset of end-to-end scholarly writing tracing the multi-month journey from initial drafts to final manuscripts.
Outcome:	The first dataset of end-to-end scholarly writing traces the multi-month journey from initial drafts to final manuscripts over four months.

Understanding Iterative Revision from Human-Written Text (2022.acl-long)

Copied to clipboard

Wanyu Du, Vipul Raheja, Dhruv Kumar, Zae Myung Kim, Melissa Lopez, Dongyeop Kang

Challenge:	This work describes IteraTeR: the first large-scale, multi-domain, edit-intention annotated corpus of iteratively revised text.
Approach:	They propose to annotate iteratively revised text using a multi-domain annotated corpus that generalizes to a variety of domains, edit intentions, revision depths, and granularities.
Outcome:	The proposed model improves automatic evaluations by integrating edit intentions with writing quality.

Balancing the Effect of Training Dataset Distribution of Multiple Styles for Multi-Style Text Transfer (2023.findings-acl)

Copied to clipboard

Debarati Das, David Ma, Dongyeop Kang

Text Generation

Challenge:	Text style transfer requires a high-quality paired dataset and quality training data.
Approach:	They propose to use a pseudo-parallel dataset to adjust the style distribution in training data to balance the style transfer model.
Outcome:	The proposed model produces more effective control effects over multiple styles than an imbalanced or skewed one.

Becoming Experienced Judges: Selective Test-Time Learning for Evaluators (2026.eacl-short)

Copied to clipboard

Seungyeon Jwa, Daechul Ahn, Reokyoung Kim, Dongyeop Kang, Jonghyun Choi

Plan ahead: Self-Supervised Text Planning for Paragraph Completion Task (2020.emnlp-main)

Challenge:	Large language models and visionlanguage models are increasingly used as automatic evaluators.
Approach:	They propose a framework that allows evaluators to improve sequentially at inference time without additional training or external signals.
Outcome:	The proposed framework outperforms strong baselines in two pairwise comparisons.

Copied to clipboard

Dongyeop Kang, Eduard Hovy

Reasoning Beyond Literal: Cross-style Multimodal Reasoning for Figurative Language Understanding (2026.findings-eacl)

Challenge:	despite the success of contextualized language models, language models cannot capture textual coherence of a long, multi-sentence document.
Approach:	They propose a paragraph completion task that predicts masked sentences in a sentence . they propose SSPlanner that predict what to say first and guides the pretrained model .
Outcome:	The proposed model outperforms baseline generation models on the paragraph completion task in automatic and human evaluation.

Copied to clipboard

Seyyed Saeid Cheshmi, Hahnemann Ortiz, James Mooney, Dongyeop Kang

Sentiment Analysis

Challenge:	figurative language is essential for expressing intent, emotion, and perspective . figural language is often dependent on Styles Reasoning, causing incongruities between expressions .
Approach:	They propose a framework that induces reasoning capabilities to compact vision–language models . figurative language is essential in expressing intent, emotion, and perspective .
Outcome:	The proposed framework can interpret multimodal figurative language, provide transparent reasoning traces, and generalize across multiple figurativ styles.

Dialogue Chain-of-Thought Distillation for Commonsense-aware Conversational Agents (2023.emnlp-main)

Copied to clipboard

Hyungjoo Chae, Yongho Song, Kai Ong, Taeyoon Kwon, Minjin Kim, Youngjae Yu, Dongha Lee, Dongyeop Kang, Jinyoung Yeo

Dialogue Systems & Conversational Agents

Challenge:	a human-like chatbot requires commonsense reasoning to comprehend and respond to information . however, identifying and aggregating key evidence within a single hop is a challenge . a knowledge distillation framework is proposed that leverages LLMs as unreliable teachers .
Approach:	They propose a framework that leverages large language models as unreliable teachers to facilitate multi-hop reasoning over a dialogue context.
Outcome:	The proposed framework leverages LLMs as unreliable teachers and selectively distills consistent and helpful rationales via alignment filters.

Strong Memory, Weak Control: An Empirical Study of Executive Functioning in LLMs (2026.eacl-long)

Copied to clipboard

Karin de Langis, Jong Inn Park, Bin Hu, Khanh Chi Le, Andreas Schramm, Michael C. Mensink, Andrew Elfenbein, Dongyeop Kang

Reasoning

Challenge:	Working memory is a critical component of human intelligence and executive functioning . it is correlated with performance on various cognitive tasks, including fluid intelligence .
Approach:	They apply working memory tasks to large language models to estimate working memory capacity . they find that LLMs exceed normative human scores, but not executive functioning benchmarks .
Outcome:	The proposed models do not show higher performance on executive functioning tasks or problem solving benchmarks.

Mary, the Cheeseburger-Eating Vegetarian: Do LLMs Recognize Incoherence in Narratives? (2026.eacl-long)

Copied to clipboard

Karin De Langis, Püren Öncel, Ryan Peters, Andrew Elfenbein, Laura Kristen Allen, Andreas Schramm, Dongyeop Kang

Challenge:	Contemporary models of (human) reading comprehension characterize comprehension as a dynamic process in which the reader continually builds and updates representations to maintain coherence and integrate new information with prior knowledge.
Approach:	They use a paired narrative dataset to examine the extent to which large language models can reliably separate incoherent and coherent stories.
Outcome:	The proposed models do not eliminate the deficits in the model internal state and behavior.

Threads of Subtlety: Detecting Machine-Generated Texts Through Discourse Motifs (2024.acl-long)

Copied to clipboard

Zae Myung Kim, Kwang Lee, Preston Zhu, Vipul Raheja, Dongyeop Kang

Challenge:	Empirical findings show that although both LLMs and humans generate distinct discourse patterns influenced by specific domains, human-written texts exhibit more structural variability, reflecting the nuanced nature of human writing in different domains.
Approach:	They propose a method to leverage hierarchical parse trees and recursive hypergraphs to uncover distinctive discourse patterns in texts written by humans and LLMs.
Outcome:	The proposed method combines hierarchical parse trees and recursive hypergraphs to uncover distinctive discourse patterns in texts produced by both LLMs and humans.

Dynamic Multi-Reward Weighting for Multi-Style Controllable Generation (2024.emnlp-main)

Copied to clipboard

Karin De Langis, Ryan Koo, Dongyeop Kang

Challenge:	Prior work explored the domain of controlled style generation, a task in which a generative language model aims to generate text with a specified style 2 . however in practice, text often contains not only a single style, but a combination of styles.
Approach:	They propose to use calibrated outputs from discriminators and dynamic weighting by discriminator gradient magnitudes to combine multiple styles in a reward function.
Outcome:	The proposed dynamic weighting outperforms static weighting approaches with respect style control while maintaining linguistic quality.

Posterior Calibrated Training on Sentence Classification Tasks (2020.acl-main)

Copied to clipboard

Taehee Jung, Dongyeop Kang, Hua Cheng, Lucas Mentch, Thomas Schaaf

Information Retrieval

Challenge:	Existing methods for posterior calibration have been used to correct poorly calibrated posterior probabilities.
Approach:	They propose a posterior calibration procedure that optimizes posterior probability distributions while minimizing calibration errors.
Outcome:	The proposed procedure reduces calibration error and improves performance on both objectives.

Tuning Large Multimodal Models for Videos using Reinforcement Learning from AI Feedback (2024.acl-long)

Copied to clipboard

Daechul Ahn, Yura Choi, Youngjae Yu, Dongyeop Kang, Jonghyun Choi

SkOTaPA: A Dataset for Skepticism Detection in Online Text after Persuasion Attempt (2024.lrec-main)

Challenge:	Recent advances in large language models have influenced the development of video large multimodal models (VLMMs).
Approach:	They propose a method that integrates video descriptions as context into a multimodal AI system to enrich the understanding of video content.
Outcome:	Empirical evaluations show that the proposed approach outperforms existing approaches for video large multimodal models (VLMMs)

Copied to clipboard

Smitha Muthya Sudheendra, Maral Abdollahi, Dongyeop Kang, Jisu Huh, Jaideep Srivastava

Sentiment Analysis

Challenge:	Persuasion attempts are a form of persuaded behavior that can be observed in various social settings, such as advertising, public health, political campaigns, and personal relationships.
Approach:	They propose to use multiple independent human annotations to detect skepticism in response to persuasion attempts on social media influencer marketing.
Outcome:	The proposed corpus detects skepticism in response to persuasion attempts on social media influencer marketing using multiple independent human annotations.

infoVerse: A Universal Framework for Dataset Characterization with Multidimensional Meta-information (2023.acl-long)

Copied to clipboard

Jaehyung Kim, Yekyung Kim, Karin de Langis, Jinwoo Shin, Dongyeop Kang

Responsible & Trustworthy NLP

Challenge:	Several methods for characterizing datasets based on model-driven meta-information have been developed, but the relationship and complementary effects of these methods have received less attention.
Approach:	They propose a framework that captures multidimensional characteristics of datasets by incorporating various model-driven meta-information.
Outcome:	The proposed framework outperforms baselines in three real-world applications and can be used in a variety of real-time problems.

Which Modality should I use - Text, Motif, or Image? : Understanding Graphs with Large Language Models (2024.findings-naacl)

Copied to clipboard

Debarati Das, Ishaan Gupta, Jaideep Srivastava, Dongyeop Kang

II-MMR: Identifying and Improving Multi-modal Multi-hop Reasoning in Visual Question Answering (2024.findings-acl)

Challenge:	Current research typically employs limited setups with small real-world graphs.
Approach:	They propose a new approach to encoding a graph with diverse modalities, such as text, image, and motif, coupled with prompts to approximate a diagram’s global connectivity.
Outcome:	The proposed approach improves performance of LLMs in graph structure analysis by focusing on homophily, motif presence, and graph difficulty.

Copied to clipboard

Jihyung Kil, Farideh Tavazoee, Dongyeop Kang, Joo-Kyung Kim

Modeling Mathematical Notation Semantics in Academic Papers (2021.findings-emnlp)

Challenge:	Existing studies have focused on assessing the model’s overall accuracy without evaluating it on different reasoning cases.
Approach:	They propose a novel idea to identify and improve multi-modal multi-hop reasoning in VQA by using two new language prompts to find a reasoning path to reach its answer.
Outcome:	The proposed model improves multi-modal multi-hop reasoning in visual question answering (VQA) it finds that the proposed model is easy to answer, simply demanding “single-hop” reasoning, whereas only a few questions require “multi-hop.”

Copied to clipboard

Hwiyeol Jo, Dongyeop Kang, Andrew Head, Marti A. Hearst