Papers by Seunghyun Yoon
MARCH: Evaluating the Intersection of Ambiguity Interpretation and Multi-hop Inference (2026.findings-acl)
Copied to clipboard
Jeonghyun Park, Ingeol Baek, Seunghyun Yoon, Haeun Jang, Aparna Garimella, Akriti Jain, Nedim Lipka, Hwanhee Lee
| Challenge: | Existing benchmarks on multi-hop QA focus on single-hop and layered ambiguity, but they focus on ambiguous questions . ambiguities can arise at any stage, complicating the reasoning process . |
| Approach: | They propose a benchmark to evaluate ambiguity in multi-hop question answering . they propose MARCH, which uses 2,209 carefully annotated questions . |
| Outcome: | The proposed framework outperforms existing approaches and significantly outperfies existing frameworks. |
Automatic Creation of Named Entity Recognition Datasets by Querying Phrase Representations (2023.acl-long)
Copied to clipboard
| Challenge: | Named entity recognition models rely on domain-specific dictionaries provided by experts . however, such dictionary sets are infeasible in many domains where they do not exist . |
| Approach: | They propose a framework that generates NER datasets with high-coverage pseudo-dictionaries . phrase retrieval models are used to retrieve popular entities rather than rare ones . |
| Outcome: | The proposed framework outperforms the previous best model by an average F1 score of 4.7 across five NER benchmark datasets. |
Multimodal Intent Discovery from Livestream Videos (2022.findings-naacl)
Copied to clipboard
Adyasha Maharana, Quan Tran, Franck Dernoncourt, Seunghyun Yoon, Trung Bui, Walter Chang, Mohit Bansal
| Challenge: | Existing models for instructional video understanding struggle to understand abstract intents . identifying procedural intent within instructional videos is a challenging task . |
| Approach: | They propose to extract instructional intent from software instructional livestreams by using a multimodal cascaded cross-attention model that integrates weaker and noisier video signals with more discriminative text signals. |
| Outcome: | The proposed model improves on baseline models and compares it to existing models. |
Propagate-Selector: Detecting Supporting Sentences for Question Answering via Graph Neural Networks (2020.lrec-1)
Copied to clipboard
| Challenge: | Existing question-answering models do not require reasoning across sentences in the given context (passage). |
| Approach: | They propose a graph neural network that propagates information over sentences to understand information that cannot be inferred when considering sentences in isolation. |
| Outcome: | The proposed approach obtains the best performance compared to the widely used answer-selection models that do not consider the intersentential relationship. |
MeetingQA: Extractive Question-Answering on Meeting Transcripts (2023.acl-long)
Copied to clipboard
| Challenge: | Meeting transcripts are a promising domain for natural language tasks . lack of annotated data impedes research on other important tasks in this domain . |
| Approach: | They propose an extractive QA dataset comprising questions asked by meeting participants and corresponding responses. |
| Outcome: | The proposed dataset extracts questions asked by meeting participants and corresponding responses from transcripts. |
A Survey on LLM-based Conversational User Simulation (2026.eacl-long)
Copied to clipboard
Bo Ni, Yu Wang, Leyao Wang, Branislav Kveton, Franck Dernoncourt, Yu Xia, Hongjie Chen, Reuben Luera, Samyadeep Basu, Subhojyoti Mukherjee, Puneet Mathur, Nesreen K. Ahmed, Junda Wu, Li Li, Huixin Zhang, Ruiyi Zhang, Tong Yu, Sungchul Kim, Jiuxiang Gu, Zhengzhong Tu, Alexa Siu, Zichao Wang, Seunghyun Yoon, Nedim Lipka, Namyong Park, Zihao Lin, Trung Bui, Yue Zhao, Tyler Derr, Ryan A. Rossi
| Challenge: | Recent advances in large language models (LLMs) have enabled high-fidelity generation of synthetic user conversation. |
| Approach: | They propose a taxonomy covering user granularity and simulation objectives . they analyze core techniques and evaluation methodologies to help them understand the latest developments . |
| Outcome: | The proposed model enables high-fidelity generation of synthetic user conversation. |
Medical Question Understanding and Answering with Knowledge Grounding and Semantic Self-Supervision (2022.coling-1)
Copied to clipboard
Khalil Mrini, Harpreet Singh, Franck Dernoncourt, Seunghyun Yoon, Trung Bui, Walter W. Chang, Emilia Farcas, Ndapa Nakashole
| Challenge: | Current medical question answering systems have difficulty processing long, detailed and informally worded questions . a growing number of approaches attempt to enhance the processing of consumer health questions - or medical question understanding . |
| Approach: | They propose a medical question understanding and answering system with knowledge grounding and semantic self-supervision that matches a user question with a trusted medical knowledge base and retrieves a fixed number of relevant sentences from the corresponding answer document. |
| Outcome: | The proposed system retrieves more relevant answers while achieving 20 times faster. |
UMIC: An Unreferenced Metric for Image Captioning via Contrastive Learning (2021.acl-short)
Copied to clipboard
| Challenge: | BERTScore and other text generation metrics do not use reference captions to evaluate image captions. |
| Approach: | They propose a new metric which does not require reference captions to evaluate image captions . they train UMIC to discriminate negative captions via contrastive learning . |
| Outcome: | The proposed metric has higher correlation than previous metrics that require multiple references. |
Hypothetical Documents or Knowledge Leakage? Rethinking LLM-based Query Expansion (2025.findings-acl)
Copied to clipboard
| Challenge: | Recent studies have demonstrated effectiveness in zero-shot retrieval tasks using large language models. |
| Approach: | They challenge this assumption by analyzing whether knowledge leakage in benchmarks contributes to performance gains. |
| Outcome: | The proposed methods have demonstrated significant performance gains across multiple benchmarks. |
PR-MCS: Perturbation Robust Metric for MultiLingual Image Captioning (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Existing image captioning metrics are vulnerable to lexical perturbations, but they are not robust to such perturbations. |
| Approach: | They propose a perturbation-robust multilingual CLIPScore which is a reference-free image captioning metric for multiple languages. |
| Outcome: | The proposed metric outperforms baseline metrics in capturing lexical noise of all various perturbation types in all five languages while maintaining a strong correlation with human judgments. |
Virtual Knowledge Graph Construction for Zero-Shot Domain-Specific Document Retrieval (2022.coling-1)
Copied to clipboard
| Challenge: | Domain-specific documents cover terminologies and specialized knowledge. |
| Approach: | They propose a domain-specific document retrieval method that embeds a document into a graph of entities and their relations into . they compare the unsupervised method with previous approaches and use it to compute relevance between queries and documents. |
| Outcome: | The proposed method outperforms baselines and fully-supervised bi-encoders in a zero-shot setting and outperformed bi-supervised approaches. |
FaithUn: Toward Faithful Forgetting in Language Models by Investigating the Interconnectedness of Knowledge (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for unlearning undesirable knowledge have overlooked complexity and interconnectedness of knowledge, authors say . previous studies have neglected the complex nature of knowledge and neglected its internal dependencies. |
| Approach: | They propose a new concept called superficial unlearning to evaluate faithfulness of unlearning in knowledge QA settings. |
| Outcome: | The proposed method shows significant effectiveness in real-world knowledge QA settings. |
PiC: A Phrase-in-Context Dataset for Phrase Understanding and Semantic Search (2023.eacl-main)
Copied to clipboard
| Challenge: | Existing benchmarks for phrase-similarity compare phrases alone (without context) and phrases with context (with or without context). |
| Approach: | They propose to use a dataset of 28K noun phrases accompanied by their contextual Wikipedia pages to train machine phrase embeddings. |
| Outcome: | The proposed dataset improves ranking-models’ accuracy and pushes span selection models near human accuracy, which is 95% Exact Match (EM) on semantic search given a query phrase and a passage. |
MACRONYM: A Large-Scale Dataset for Multilingual and Multi-Domain Acronym Extraction (2022.coling-1)
Copied to clipboard
Amir Pouran Ben Veyseh, Nicole Meister, Seunghyun Yoon, Rajiv Jain, Franck Dernoncourt, Thien Huu Nguyen
| Challenge: | Acronym extraction is the task of identifying acronyms and their expanded forms in texts . existing AE methods for English are limited to specific languages and domains . |
| Approach: | They propose to annotate 27,200 sentences in 6 different languages and 2 new domains for AE. |
| Outcome: | The proposed dataset shows that AE in different languages and learning settings has unique challenges . |
KPQA: A Metric for Generative Question Answering Using Keyphrase Weights (2021.naacl-main)
Copied to clipboard
| Challenge: | Existing n-gram similarity metrics fail to discriminate the incorrect answers due to the free-form of the answer. |
| Approach: | They propose a new metric that assigns different weights to each token via keyphrase prediction to judge the correctness of GenQA. |
| Outcome: | The proposed metric has a significantly higher correlation with human judgments than existing metrics in various datasets. |
From Selection to Generation: A Survey of LLM-based Active Learning (2025.acl-long)
Copied to clipboard
Yu Xia, Subhojyoti Mukherjee, Zhouhang Xie, Junda Wu, Xintong Li, Ryan Aponte, Hanjia Lyu, Joe Barrow, Hongjie Chen, Franck Dernoncourt, Branislav Kveton, Tong Yu, Ruiyi Zhang, Jiuxiang Gu, Nesreen K. Ahmed, Yu Wang, Xiang Chen, Hanieh Deilamsalehy, Sungchul Kim, Zhengmian Hu, Yue Zhao, Nedim Lipka, Seunghyun Yoon, Ting-Hao Kenneth Huang, Zichao Wang, Puneet Mathur, Soumyabrata Pal, Koyel Mukherjee, Zhehao Zhang, Namyong Park, Thien Huu Nguyen, Jiebo Luo, Ryan A. Rossi, Julian McAuley
| Challenge: | Large Language Models (LLMs) have been used for selection and training of data for active learning. |
| Approach: | They propose an intuitive taxonomy that categorizes LLM-based active learning techniques and discuss the transformative roles they can play in the active learning loop. |
| Outcome: | The proposed model can generate entirely new data instances and provide more cost-effective annotations with fewer labeled data instances. |
Can Large Language Models Keep Up? Benchmarking Online Adaptation to Continual Knowledge Streams (2026.acl-long)
Copied to clipboard
Jiyeon Kim, Hyunji Lee, Dylan Zhou, Sue Hyun Park, Seunghyun Yoon, Trung Bui, Franck Dernoncourt, Sungmin Cha, Minjoon Seo
| Challenge: | Existing models and agentic memory systems fail to adapt robustly to OAKS, demonstrating delays in state-tracking and susceptibility to distraction within streaming environments. |
| Approach: | They propose a benchmark to evaluate models' ability to adapt to changing knowledge over streaming . they use two datasets to analyze how facts evolve over time . |
| Outcome: | The proposed benchmark evaluates models in an online adaptation setting over streaming, continually updating knowledge. |
Keyphrase Prediction from Video Transcripts: New Dataset and Directions (2022.coling-1)
Copied to clipboard
Amir Pouran Ben Veyseh, Quan Hung Tran, Seunghyun Yoon, Varun Manjunatha, Hanieh Deilamsalehy, Rajiv Jain, Trung Bui, Walter W. Chang, Franck Dernoncourt, Thien Huu Nguyen
| Challenge: | Existing studies on keyphrase prediction have focused on formal texts and informal-text domains. |
| Approach: | They propose to annotate large-scale video transcripts with keyphrases from live-stream video . they propose to feed models with paragraph-level keyphrase extraction to foster future research . |
| Outcome: | The proposed model improves keyphrase prediction in live-stream video transcripts by feeding models with paragraph-level keyphrases. |
PDFTriage: Question Answering over Long, Structured Documents (2024.emnlp-industry)
Copied to clipboard
| Challenge: | Existing approaches to document QA use a pre-retrieval step to retrieve the relevant context from documents, but this is incongruous with the user's mental model of the document. |
| Approach: | They propose an approach called PDFTriage that enables models to retrieve the context based on either structure or content. |
| Outcome: | The proposed approach can retrieve context based on structure or content across several classes of questions where existing retrieval-augmented LLMs fail. |
Towards Enhancing Coherence in Extractive Summarization: Dataset and Experiments with LLMs (2024.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for extractive summarization lack coherence, despite improvements . a human-annotated dataset is used to improve coherency of extractive summary . |
| Approach: | They propose to use human-annotated datasets to create coherent extractive summaries . they use supervised fine-tuning and natural language user feedback to enhance coherence . |
| Outcome: | The proposed dataset shows that LLMs can produce coherent summaries with human feedback. |
Few-Shot Intent Detection via Contrastive Pre-Training and Fine-Tuning (2021.emnlp-main)
Copied to clipboard
Jianguo Zhang, Trung Bui, Seunghyun Yoon, Xiang Chen, Zhiwei Liu, Congying Xia, Quan Hung Tran, Walter Chang, Philip Yu
| Challenge: | Existing methods address few-shot intent detection tasks from two perspectives: data augmentation and task-adaptive training with pre-trained models. |
| Approach: | They propose a few-shot intent detection schema using contrastive pre-training and fine-tuning. |
| Outcome: | The proposed method achieves state-of-the-art performance on three challenging intent detection datasets under 5-shot and 10-shot settings. |
VLind-Bench: Measuring Language Priors in Large Vision-Language Models (2025.findings-naacl)
Copied to clipboard
| Challenge: | Large Vision-Language Models suffer from a problem known as language prior . such language priors can lead to undesirable biases and hallucinations when dealing with images that are out of distribution. |
| Approach: | They propose a benchmark to measure the language priors of Large Vision-Language Models. |
| Outcome: | The proposed benchmark is the first specifically designed to measure the language priors, or blindness, of LVLMs. |
Generating Diverse Hypotheses for Inductive Reasoning (2025.naacl-long)
Copied to clipboard
| Challenge: | Recent studies suggest that large language models (LLMs) can engage in inductive reasoning by sampling multiple hypotheses about the rules and selecting the one that best explains the observations. |
| Approach: | They propose to increase the temperature parameter to enhance diversity by sampling multiple hypotheses and selecting the one that best explains the observations. |
| Outcome: | The proposed method improves diversity while maintaining text quality while increasing temperature. |
PEEB: Part-based Image Classifiers with an Explainable and Editable Language Bottleneck (2024.findings-naacl)
Copied to clipboard
| Challenge: | CLIP-based classifiers rely on the prompt containing a class name that is known to the text encoder and perform poorly on new classes or the classes whose names rarely appear on the Internet. |
| Approach: | They propose to use a set of text descriptors to express a class name into a textual descriptable and match the embeddings of the detected parts to their textual ones to compute a logit score. |
| Outcome: | The proposed classifier outperforms CLIP-based classifiers on zero-shot and supervised learning settings by 88.80% and 92.20% accuracy on CUB-200 and Stanford Dogs-120. |
CORG: Generating Answers from Complex, Interrelated Contexts (2025.naacl-long)
Copied to clipboard
| Challenge: | Existing approaches to analyzing knowledge in a corpus often focus on single factors in isolation. |
| Approach: | They propose a framework that organizes multiple contexts into independently processed groups . they classify these relationships into distracting, ambiguous, counterfactual, and duplicated . |
| Outcome: | The proposed framework outperforms existing grouping methods and single-context approaches. |
Fast and Accurate Deep Bidirectional Language Representations for Unsupervised Learning (2020.acl-main)
Copied to clipboard
| Challenge: | Existing deep bidirectional language models are limited by repetitive inferences on unsupervised tasks for the computation of contextual language representations. |
| Approach: | They propose a deep bidirectional language model called a Transformer-based Text Autoencoder (T-TA) it computes contextual language representations without repetition and shows competitive or even better accuracies than BERT . |
| Outcome: | The proposed model performs six times faster on a reranking task and twelve times faster in a semantic similarity task. |
FIFA: Unified Faithfulness Evaluation Framework for Text-to-Video and Video-to-Text Generation (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing evaluation methods for VideoMLLMs are limited to one task and fail to assess hallucinations in open-ended, free-form responses. |
| Approach: | They propose a unified framework that extracts comprehensive descriptive facts and models their semantic dependencies via a Spatio-Temporal Semantic Dependency Graph. |
| Outcome: | The proposed framework aligns more closely with human judgment than existing evaluation methods and improves factual consistency in both text and video generation. |
Simple Questions Generate Named Entity Recognition Datasets (2022.emnlp-main)
Copied to clipboard
| Challenge: | Recent named entity recognition models rely on human-annotated datasets . however, in-domain dictionaries and sentences are often unavailable or expensive to construct for many entity types. |
| Approach: | They propose an ask-to-generate approach which automatically generates NER datasets by asking natural language questions to an open-domain question answering system. |
| Outcome: | The proposed model outperforms the previous best model by 19.5 F1 score on six benchmarks and achieves state-of-the-art performance. |
FIZZ: Factual Inconsistency Detection by Zoom-in Summary and Zoom-out Document (2024.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for evaluating factual consistency in abstractive summarization systems have significant limitations, especially on refinement and interpretability. |
| Approach: | They propose a method for detecting summary factual inconsistency based on fine-grained atomic facts decomposition and adaptive granularity expansion. |
| Outcome: | The proposed method outperforms existing systems on the AGGREFACT benchmark dataset and achieves state-of-the-art performance. |
Fine-grained Image Captioning with CLIP Reward (2022.findings-naacl)
Copied to clipboard
| Challenge: | Modern image captioning models are usually trained with text similarity objectives . reference captions often describe only the most salient objects in images . |
| Approach: | They propose to use CLIP to calculate multi-modal similarity and use it as a reward function . they propose a simple finetuning strategy to improve grammar that does not require extra text annotation. |
| Outcome: | The proposed model generates more distinctive captions than the CIDEroptimized model on text-to-image retrieval and fineCapEval. |
Fine-tuning CLIP Text Encoders with Two-step Paraphrasing (2024.findings-eacl)
Copied to clipboard
| Challenge: | Contrastive language-image pre-training models have demonstrated considerable success across various vision-language tasks, such as text-to-image retrieval. |
| Approach: | They propose a fine-tuning approach to enhance the representations of CLIP models for paraphrases by leveraging large language models. |
| Outcome: | The proposed model improves on baseline models across paraphrased retrieval, visual genome relation and attribution, and seven semantic textual similarity tasks. |
Learning to Rank Question-Answer Pairs Using Hierarchical Recurrent Encoder with Latent Topic Clustering (N18-1)
Copied to clipboard
| Challenge: | Existing models for sentence pair ranking are based on hierarchical recurrent neural network and latent topic clustering module. |
| Approach: | They propose a hierarchical recurrent neural network and latent topic clustering module to adapt a recursive hierarchic neural network to rank candidate answers. |
| Outcome: | The proposed model shows small performance degradations in longer text comprehension compared to current models which suffer from it. |
Instruction Tuning with and without Context: Behavioral Shifts and Downstream Impact (2026.eacl-long)
Copied to clipboard
Hyunji Lee, Seunghyun Yoon, Yunjae Won, Hanseok Oh, Geewook Kim, Trung Bui, Franck Dernoncourt, Elias Stengel-Eskin, Mohit Bansal, Minjoon Seo
| Challenge: | Prior work on instruction tuning datasets combined these data types without examining their distinct effects. |
| Approach: | They investigate how training LLMs with or without context affects model behavior and performance . they find that using context-augmented data as the backbone for vision-language models reduces hallucination . |
| Outcome: | The proposed training with context-augmented data reduces hallucination and improves grounding in the visual domain. |
GUI Agents: A Survey (2025.findings-acl)
Copied to clipboard
Dang Nguyen, Jian Chen, Yu Wang, Gang Wu, Namyong Park, Zhengmian Hu, Hanjia Lyu, Junda Wu, Ryan Aponte, Yu Xia, Xintong Li, Jing Shi, Hongjie Chen, Viet Dac Lai, Zhouhang Xie, Sungchul Kim, Ruiyi Zhang, Tong Yu, Mehrab Tanjim, Nesreen K. Ahmed, Puneet Mathur, Seunghyun Yoon, Lina Yao, Branislav Kveton, Jihyung Kil, Thien Huu Nguyen, Trung Bui, Tianyi Zhou, Ryan A. Rossi, Franck Dernoncourt
| Challenge: | Large Foundation Models (LFMs) have transformed the landscape of AI research and day-to-day life. |
| Approach: | They propose a framework that delineates GUI agents' perception, reasoning, planning, and acting capabilities. |
| Outcome: | The proposed framework delineates their perception, reasoning, planning, and acting capabilities. |
SlimLM: An Efficient Small Language Model for On-Device Document Assistance (2025.acl-demo)
Copied to clipboard
| Challenge: | Small language models (SLMs) show promise for mobile deployment, but their real world performance and applications on smartphones remain understudied. |
| Approach: | They propose a slim language model with a model size of 125M to 8B and a context length of 8B for efficient on-device processing. |
| Outcome: | The proposed model is based on a Samsung Galaxy S24 and shows comparable or superior performance. |
A Gradually Soft Multi-Task and Data-Augmented Approach to Medical Question Understanding (2021.acl-long)
Copied to clipboard
Khalil Mrini, Franck Dernoncourt, Seunghyun Yoon, Trung Bui, Walter Chang, Emilia Farcas, Ndapa Nakashole
| Challenge: | Existing methods for medical question understanding often fail to provide high recall in answer retrieval. |
| Approach: | They propose a multi-task learning method with data augmentation for medical question understanding that uses just one dataset to optimize for both tasks. |
| Outcome: | The proposed method outperforms existing MTL methods across 4 datasets of medical question pairs in ROUGE scores, RQE accuracy and human evaluation. |
Assessing News Thumbnail Representativeness: Counterfactual text can enhance the cross-modal matching ability (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing models for visual and language processing struggle to match news actors’ visual and textual appearances. |
| Approach: | They propose a method that generates counterfactual news thumbnail images and text pairs to assess whether a news thumbnail image represents the actors discussed in the news text. |
| Outcome: | The proposed method can boost the performance for assessing news thumbnail representativeness, supporting the hypothesis. |
Offensive Content Detection via Synthetic Code-Switched Text (2022.coling-1)
Copied to clipboard
| Challenge: | Existing methods to detect offensive content in social media platforms are limited by the availability of labeled code-switched data. |
| Approach: | They propose a method for generating synthetic code-switched offensive content data using human-generated data and a keyword classification baseline. |
| Outcome: | The proposed algorithm can be used to generate synthetic code-switched offensive content data and train it on human-generated data. |
QACE: Asking Questions to Evaluate an Image Caption (2021.findings-emnlp)
Copied to clipboard
| Challenge: | Existing metric for image captioning evaluation is based on n-gram similarity metrics but these fail to capture semantic errors in captions. |
| Approach: | They propose a new metric based on Question Answering for Caption Evaluation to evaluate image captioning based upon Question Generation and Question Answers systems. |
| Outcome: | The proposed metric is multi-modal, reference-less and explainable. |