Papers by Dongha Lee
OoMMix: Out-of-manifold Regularization in Contextual Embedding Space for Text Classification (2021.acl-long)
Copied to clipboard
| Challenge: | Recent studies on neural networks with pre-trained weights focus on low-dimensional subspace where the embedding vectors computed from input words are located. |
| Approach: | They propose an approach to find and regularize the remainder of the space, referred to as out-of-manifold, which cannot be accessed through the words. |
| Outcome: | The proposed approach is able to fine-tune the out-of-manifold embedding space on text classification benchmarks. |
TransferCVLM: Transferring Cross-Modal Knowledge for Vision-Language Modeling (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Recent large vision-language multimodal models pre-trained with huge amount of image-text pairs show remarkable performances in downstream tasks. |
| Approach: | They propose a method of efficient knowledge transfer that integrates pre-trained uni-modal models into a combined vision-language model without pre-training . they propose to fine-tune the model and transfer multimodal knowledge from a teacher vision-linguistic model to the CVLM for each task application. |
| Outcome: | The proposed method outperforms existing vision-language models in downstream tasks. |
Commonsense-augmented Memory Construction and Management in Long-term Conversations via Context-aware Persona Refinement (2024.eacl-short)
Copied to clipboard
| Challenge: | Memorizing and utilizing speakers’ personas is a common practice for response generation in long-term conversations, yet human-authored datasets often provide uninformative persona sentences that hinder response quality. |
| Approach: | They propose a framework that leverages commonsense-based persona expansion to address such issues in long-term conversations. |
| Outcome: | The proposed framework facilitates better response generation via human-like persona refinement. |
Extracting Chemical-Protein Interactions via Calibrated Deep Neural Network and Self-training (2020.findings-emnlp)
Copied to clipboard
| Challenge: | Several natural language processing methods have been used to extract interactions between chemicals and proteins from biomedical text data. |
| Approach: | They propose a method to extract chemical–protein interactions from biomedical text data . they use a pre-trained language-understanding model and calibration techniques to estimate uncertainty . |
| Outcome: | The proposed approach achieves state-of-the-art performance on the Biocreative VI ChemProt task while preserving higher calibration abilities. |
Can Code-Switched Texts Activate a Knowledge Switch in LLMs? A Case Study on English-Korean Code-Switching (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Recent large language models (LLMs) demonstrate multilingual abilities, yet they are English-centric due to dominance of English in training corpora. |
| Approach: | They propose to use a synthetic English-korean CS question-answering dataset to investigate this potential. |
| Outcome: | The proposed model can activate, identify and leverage knowledge for reasoning in low-resource languages. |
Pearl: A Review-driven Persona-Knowledge Grounded Conversational Recommendation Dataset (2024.findings-acl)
Copied to clipboard
Minjin Kim, Minju Kim, Hana Kim, Beong-woo Kwak, SeongKu Kang, Youngjae Yu, Jinyoung Yeo, Dongha Lee
| Challenge: | Existing datasets for conversational recommender systems lack specific user preferences and explanations for recommendations . current datasets lack specific preferences, hindering high-quality recommendations despite advances in large language models . |
| Approach: | They propose to synthesize a conversational recommendation dataset with persona- and knowledge-augmented LLM simulators to address these challenges. |
| Outcome: | The proposed dataset outperforms baselines in human and automatic evaluations. |
MT-RAIG: Novel Benchmark and Evaluation Framework for Retrieval-Augmented Insight Generation over Multiple Tables (2025.acl-long)
Copied to clipboard
| Challenge: | Existing studies on table-based reasoning focus on a single gold table, not multiple tables . a persistent demand for robust table understanding systems is resulting from the complexity of table data . |
| Approach: | They propose a MT-RAIG Bench to evaluate systems on Retrieval-Augmented Insight Generation over Mulit-Tables. |
| Outcome: | The proposed framework improves human quality judgments on the generated insights. |
Exploring Language Model’s Code Generation Ability with Auxiliary Functions (2024.findings-naacl)
Copied to clipboard
| Challenge: | Auxiliary function is a useful component to improve language model’s code generation ability, but a systematic exploration of how they affect has yet to be done. |
| Approach: | They construct a human-crafted evaluation set which contains examples of two functions where one function assists the other to examine their ability in a multifaceted way. |
| Outcome: | The proposed model is underutilized to call the auxiliary function, suggesting future directions to enhance their implementation by eliciting the supplementary function call ability encoded in the models. |
GENDEX: Generative Data Augmentation Strategy Leveraging External Data for Abstractive Dialogue Summarization (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing methods to summarize text data are limited by the lack of data. |
| Approach: | They propose a method that uses external data to generate synthetic dialogues from short texts containing people and their interpersonal interactions. |
| Outcome: | The proposed method shows robust performance, generalizability, and scalability regardless of complexity of dialogues. |
Evidence-Focused Fact Summarization for Knowledge-Augmented Zero-Shot Question Answering (2024.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for enhancing QA performance of Large Language Models (LLMs) have limitations, including duplicated entities or relations, reduced evidence density, and failure to highlight crucial evidence. |
| Approach: | They propose an Evidence-focused Fact Summarization framework for enhanced QA with knowledge-augmented Large Language Models (LLMs) that incorporates external knowledge into LLMs to improve QA performance. |
| Outcome: | The proposed framework improves LLM’s zero-shot QA performance especially when noisy facts are retrieved. |
Taxonomy-guided Semantic Indexing for Academic Paper Search (2024.emnlp-main)
Copied to clipboard
| Challenge: | Academic paper search often struggles to match underlying academic concepts between queries and documents. |
| Approach: | They propose a framework that extracts key concepts from papers and organizes them as a semantic index guided by an academic taxonomy. |
| Outcome: | The proposed framework can be flexibly employed to enhance existing retrieval frameworks. |
Cactus: Towards Psychological Counseling Conversations using Cognitive Behavioral Theory (2024.findings-emnlp)
Copied to clipboard
Suyeon Lee, Sunghwan Kim, Minju Kim, Dongjin Kang, Dongil Yang, Harim Kim, Minseok Kang, Dayi Jung, Min Kim, Seungbeen Lee, Kyong-Mee Chung, Youngjae Yu, Dongha Lee, Jinyoung Yeo
| Challenge: | Existing models that use large language models are not available due to ethical concerns, and data privacy concerns are a concern. |
| Approach: | They propose a multi-turn dialogue dataset that emulates real-life counseling interactions using the goal-oriented approach of Cognitive Behavioral Therapy (CBT). |
| Outcome: | The proposed model outperforms other models in counseling skills, highlighting its effectiveness and potential as a counseling agent. |
Early Stopping Based on Unlabeled Samples in Text Classification (2022.acl-long)
Copied to clipboard
| Challenge: | Existing methods to stop models from overfitting are based on a separate validation set, but in low resource settings, a small validation set may not be representative enough. |
| Approach: | They propose a method that uses unlabeled samples to estimate the class distribution of the unlabed samples. |
| Outcome: | The proposed method performs better than existing stop-methods in balanced and imbalanced data settings. |
Stop Playing the Guessing Game! Evaluating Conversational Recommender Systems via Target-free User Simulation (2025.findings-emnlp)
Copied to clipboard
| Challenge: | despite advances in CRSs, reliably assessing their ability to elicit preferences remains a challenge. |
| Approach: | They propose a user-CRS evaluation protocol with target-free user simulators . they show that current evaluation metrics emphasize single-turn recall of target items . |
| Outcome: | The proposed evaluation protocol is based on a simulation-based evaluation environment. |
Can Large Language Models be Effective Online Opinion Miners? (2025.emnlp-main)
Copied to clipboard
| Challenge: | OOMB is a novel benchmark designed to assess the ability of large language models (LLMs) to extract and analyze opinions from diverse and complex online environments. |
| Approach: | They propose an online opinion mining benchmark to assess the ability of large language models to extract and analyze opinions from diverse online environments. |
| Outcome: | The proposed benchmark assesses the ability of large language models to mine opinions effectively from diverse and complex online environments. |
Toward Interpretable Semantic Textual Similarity via Optimal Transport-based Contrastive Sentence Learning (2022.acl-long)
Copied to clipboard
| Challenge: | Existing studies have shown that a pretrained language model can capture sentence similarity but there is no interpretation method for the sentence similarities. |
| Approach: | They propose a pretrained language model that captures sentence similarity between embeddings and a transport-based distance measure that leverages semantically-aligned token pairs. |
| Outcome: | The proposed framework outperforms baselines on both STS and interpretable-STS benchmarks and provides interpretation consistent with human judgement. |
Unveiling Implicit Table Knowledge with Question-Then-Pinpoint Reasoner for Insightful Table Summarization (2024.findings-emnlp)
Copied to clipboard
| Challenge: | a novel table reasoning framework is needed to uncover the hidden knowledge hidden within the explicit table cells. |
| Approach: | They propose a table reasoning framework Question-then-pinpoint that can self-question table knowledge and answer it faithfully. |
| Outcome: | The proposed framework can self-question and answer the knowledge by pinpointing evidence on the table. |
VerifiNER: Verification-augmented NER via Knowledge-grounded Reasoning with Large Language Models (2024.acl-long)
Copied to clipboard
| Challenge: | Recent approaches in domain-specific named entity recognition (NER) have shown remarkable advances, but they still lack faithfulness, producing erroneous predictions. |
| Approach: | They propose a framework that revises errors from existing NER methods using knowledge to produce more faithful predictions. |
| Outcome: | The proposed framework can validate errors from existing models as a model-agnostic approach. |
Eliciting Instruction-tuned Code Language Models’ Capabilities to Utilize Auxiliary Function for Code Generation (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Using auxiliary functions to implement functions is important for instruction-tuned models because it reduces the implementation difficulty of a target function compared to implementing them from scratch. |
| Approach: | They propose several ways to provide auxiliary functions to the models by adding them to the query or providing a response prefix to incorporate the ability to utilize auxiliary function with the instruction following capability. |
| Outcome: | The proposed models outperform the recent powerful language models, gpt-4o, in the code generation task. |
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
| 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. |
RTSUM: Relation Triple-based Interpretable Summarization with Multi-level Salience Visualization (2024.naacl-demo)
Copied to clipboard
| Challenge: | Abstractive summarization has emerged as a critical tool in the era of information overload. |
| Approach: | They propose an unsupervised summarization framework that utilizes relation triples as the basic unit for summarizing. |
| Outcome: | The proposed framework visualizes salience levels for sentences, relation triples, and phrases. |
Evidentiality-aware Retrieval for Overcoming Abstractiveness in Open-Domain Question Answering (2024.findings-eacl)
Copied to clipboard
| Challenge: | Existing approaches to ODQA use a simple yet effective retriever-reader framework, but this approach is not always effective in abstractive tasks. |
| Approach: | They propose a method that leverages synthetic distractor samples to learn to discriminate evidence passages from distractors. |
| Outcome: | The proposed method is validated on multiple abstractive open-domain question answering tasks. |
RoToR: Towards More Reliable Responses for Order-Invariant Inputs (2025.acl-long)
Copied to clipboard
| Challenge: | Existing solutions to positional bias in listwise inputs are limited on practical problems . e.g., lost-in-the-middle problem is a common problem for listwise models . |
| Approach: | They propose a zero-shot order-invariant LM for order- invariant inputs with minimal modifications of positional IDs and Selective Routing for listwise tasks. |
| Outcome: | The proposed framework can handle order-invariant and sensitive inputs in listwise tasks. |
Topic Taxonomy Expansion via Hierarchy-Aware Topic Phrase Generation (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods for topic taxonomies focus on frequent terms and local topic-subtopic relations, which leads to limited topic term coverage. |
| Approach: | They propose a framework for topic taxonomy expansion that directly generates topic-related terms belonging to new topics. |
| Outcome: | The proposed framework outperforms baseline methods on two real-world text corpora. |
Self-Consistent Reasoning-based Aspect-Sentiment Quad Prediction with Extract-Then-Assign Strategy (2024.findings-acl)
Copied to clipboard
| Challenge: | Recent studies have developed powerful generative methods for aspect sentiment quad prediction (ASQP) but they still suffer from imprecise predictions and limited interpretability due to data scarcity and inadequate modeling of the quadruplet composition process. |
| Approach: | They propose a self-consistent reasoning-based aspect sentiment quadruple prediction framework which generates reasonings and corresponding quadruples in sequence. |
| Outcome: | The proposed model significantly improves its ability to handle complex reasoning tasks and correctly predict quadruplets through consistency voting, resulting in enhanced interpretability and accuracy in aspect sentiment quadr uplp prediction. |
Towards Personalized Conversational Sales Agents: Contextual User Profiling for Strategic Action (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Conversational Recommender Systems (CRSs) aim to engage users in dialogue to provide tailored recommendations. |
| Approach: | They propose a task that integrates preference elicitation, recommendation and persuasion . they propose CSALES, a conversational sales agent that infers contextual user profiles . |
| Outcome: | The proposed task integrates preference elicitation, recommendation, and persuasion within a conversational framework. |
Towards Lifelong Dialogue Agents via Timeline-based Memory Management (2025.naacl-long)
Copied to clipboard
Kai Tzu-iunn Ong, Namyoung Kim, Minju Gwak, Hyungjoo Chae, Taeyoon Kwon, Yohan Jo, Seung-won Hwang, Dongha Lee, Jinyoung Yeo
| Challenge: | Existing studies focus on getting rid of outdated memories to improve retrieval quality, but we argue that such memories provide rich, important contextual cues for response generation (RG). |
| Approach: | They propose a framework for LLM-based lifelong dialogue agents that discards memory removal and manages large-scale memories by linking them based on their temporal and cause-effect relation. |
| Outcome: | The proposed framework augments RG with memory timelines based on evolution or causality of relevant past events. |
Do LLMs Have Distinct and Consistent Personality? TRAIT: Personality Testset designed for LLMs with Psychometrics (2025.findings-naacl)
Copied to clipboard
Seungbeen Lee, Seungwon Lim, Seungju Han, Giyeong Oh, Hyungjoo Chae, Jiwan Chung, Minju Kim, Beong-woo Kwak, Yeonsoo Lee, Dongha Lee, Jinyoung Yeo, Youngjae Yu
| Challenge: | Recent advances in Large Language Models (LLMs) have led to their adaptation as conversational agents. |
| Approach: | They propose a new benchmark that uses 8K multi-choice questions to assess the personality of Large Language Models. |
| Outcome: | The proposed personality test outperforms existing personality tests for LLMs in reliability and validity. |
Make Compound Sentences Simple to Analyze: Learning to Split Sentences for Aspect-based Sentiment Analysis (2024.findings-emnlp)
Copied to clipboard
| Challenge: | generative methods have shown promising results for extracting sentiment quadruplets . compound sentences can contain multiple quadroutlets, making extraction difficult . |
| Approach: | They propose an Aspect Term Oriented Sentence Splitter which simplifies compound sentences into simpler and clearer forms. |
| Outcome: | The proposed method outperforms existing methods in ASQP and ACOS tasks. |
P-Check: Advancing Personalized Reward Model via Learning to Generate Dynamic Checklist (2026.acl-long)
Copied to clipboard
| Challenge: | Existing approaches to personalized reward modeling treat user context as static or implicit conditioning signal, failing to capture dynamic nature of human judgment. |
| Approach: | They propose a personalized reward modeling framework that synthesizes dynamic evaluation criteria for guiding the reward prediction. |
| Outcome: | The proposed framework improves reward accuracy and enhances downstream personalized generation. |
Why These Documents? Explainable Generative Retrieval with Hierarchical Category Paths (2026.findings-acl)
Copied to clipboard
| Challenge: | Generative retrieval directly decodes a document identifier, making it impossible to provide explanations for its retrieval decision. |
| Approach: | They propose a hierarchical category path-Enhanced Generative Retrieval that generates category paths step-by-step and decodes docid. |
| Outcome: | The proposed method provides explanations for retrieval decision by generating hierarchical category paths step-by-step and decoding docid. |
Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization (2025.acl-long)
Copied to clipboard
| Challenge: | Existing benchmarks for reward models show a weak correlation with performance of optimized policies . existing benchmarks do not accurately assess the true capabilities of reward models . |
| Approach: | They explore how reward overoptimization captures how well a reward model aligns with human preferences and the dynamics of the learning signal it provides to the policy. |
| Outcome: | The proposed benchmarks show that reward overoptimization is a weak factor . the high correlation with degree of overoptimalization leads to lower correlation with downstream performance . |
How Diversely Can Language Models Solve Problems? Exploring the Algorithmic Diversity of Model-Generated Code (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Language models (LMs) have exhibited impressive abilities in generating code from natural language requirements. |
| Approach: | They propose to introduce various metrics with inter-code similarity to evaluate the diversity of generated code by comparing model-generated solutions with human-written ones. |
| Outcome: | The proposed method leverages LMs’ capabilities in code understanding and reasoning, resulting in a set of metrics that represent the number of algorithms in model-generated solutions. |
Domain Knowledge Transferring for Pre-trained Language Model via Calibrated Activation Boundary Distillation (2022.acl-long)
Copied to clipboard
| Challenge: | Pretrained language models are used to boost their performance on downstream tasks . pretraining with in-domain texts requires considerable in- domain data and training resources . |
| Approach: | They propose a domain knowledge transferring framework for pre-trained language models without additional in-domain pretraining. |
| Outcome: | The proposed framework extracts domain knowledge from an existing in-domain pretrained language model and transfers it to other PLMs by applying knowledge distillation. |