Recent advances in large language models (LLMs) have opened new possibilities for table-based tasks. However, most existing methods remain confined to single-table settings, limiting their applicability to real-world databases composed of multiple interrelated tables. In multi-table scenarios, LLMs face two key challenges: reasoning over relational structures beyond sequential text, and handling the input length limitations imposed by large-scale table concatenation. To address these issues, we propose Guided Relational Integration for multiple Tables (GRIT), a lightweight method that converts relational schemas into LLM-friendly textual representations. GRIT employs hashing-based techniques to efficiently infer primary–foreign key relationships and constructs prompts that explicitly encode relevant join paths and question-relevant columns. When applied to off-the-shelf LLMs, GRIT consistently improves table-column retrieval performance across diverse multi-table benchmarks while significantly reducing memory and computational overhead.
@inproceedings{kang-etal-2025-grit,title={{GRIT}: Guided Relational Integration for Efficient Multi-Table Understanding},author={Kang, Yujin and Woo, Park Seong and Cho, Yoon-Sik},booktitle={Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing},month=nov,year={2025},address={Suzhou, China},publisher={Association for Computational Linguistics},url={https://aclanthology.org/2025.emnlp-main.1118/},doi={10.18653/v1/2025.emnlp-main.1118},pages={21984--21997},}
CCS
Can Personal Health Information Be Secured in LLM? Privacy Attack and Defense in the Medical Domain
Yujin Kang, Eunsun Kim , and Yoon-Sik Cho
In Proceedings of the 2025 ACM SIGSAC Conference on Computer and Communications Security , Oct 2025
Recent advancements have shown that Large Language Models (LLMs) possess significant versatility, making them suitable for applications in many areas. Several studies have shown how general-purpose LLMs can be adapted to domain-specific tasks. However, these domain-adapted LLMs can be exposed to greater privacy risks, which are especially exacerbated in the medical field. In this paper, we present the study investigating the susceptibility of LLMs to leaking sensitive health information. We conduct prompt-based attacks on LLMs trained with medical datasets, showing that medical LLMs can inadvertently disclose confidential patient data. To contribute towards mitigating privacy risks in the medical domain, we implement red teaming defense strategies to make LLMs robust against malicious attacks. For this medical red teaming approach, we develop and publicly release MediRed, a dataset of 1,000 red team attacks. By leveraging this dataset to enhance our defense mechanisms, we achieve up to 56% improvement in privacy protection compared to base models.
@inproceedings{kang-kim-cho-2025-private-med,title={Can Personal Health Information Be Secured in LLM? Privacy Attack and Defense in the Medical Domain},author={Kang, Yujin and Kim, Eunsun and Cho, Yoon-Sik},booktitle={Proceedings of the 2025 ACM SIGSAC Conference on Computer and Communications Security},month=oct,year={2025},publisher={Association for Computing Machinery},address={Taipei, Taiwan},url={https://doi.org/10.1145/3719027.3765105},doi={10.1145/3719027.3765105},pages={4199--4213},presentation={Oral Presentation}}
WWW
Leveraging Refined Negative Feedback with LLM for Recommender Systems
Chanwoo Jeong , Yujin Kang, and Yoon-Sik Cho
In Companion Proceedings of the ACM on Web Conference 2025 , Apr 2025
Recently, there has been growing research on negative feedback in recommender systems. These studies use a fixed threshold to binarize feedback into positive or negative. However, such an approach bears limitations when the rating habits for expressing disappointment differ across users or when ratings are noisy. Motivated by the remarkable success of Large Language Models (LLMs), we investigate how LLM can address this challenge on the fly. To this end, we present ReFINe, Resurrecting Falsely Identified Negative feedback with LLM. ReFINe classifies the negative feedback into two distinct types: Falsely identified negative with positive signals and Confirmed negative with only negative signals. To the best of our knowledge, our work is the first to propose and demonstrate the distinction between two perspectives on negative feedback.
@inproceedings{jeong-kang-cho-2025-refine,title={Leveraging Refined Negative Feedback with LLM for Recommender Systems},author={Jeong, Chanwoo and Kang, Yujin and Cho, Yoon-Sik},booktitle={Companion Proceedings of the ACM on Web Conference 2025},month=apr,year={2025},publisher={Association for Computing Machinery},address={Sydney NSW, Australia},url={https://doi.org/10.1145/3701716.3715538},doi={10.1145/3701716.3715538},pages={1028--1032},}
AAAI
Beyond Single Emotion: Multi-label Approach to Conversational Emotion Recognition
Yujin Kang, and Yoon-Sik Cho
Proceedings of the AAAI Conference on Artificial Intelligence, Feb 2025
Emotion recognition in conversation (ERC) has been promoted with diverse approaches in the recent years. However, many studies have pointed out that emotion shift and confusing labels make it difficult for models to distinguish between different emotions. Existing ERC models suffer from these problems when the emotions are forced to be mapped into single label. In this paper, we utilize our strategies for extending single label to multi-labels. We then propose a multi-label classification framework for emotion recognition in conversation (ML-ERC). Specifically, we introduce weighted supervised contrastive learning tailored for multi-label, which can easily applied to previous ERC models. The empirical results on existing task with single label support the efficacy of our approach, which is more effective in the most challenging settings: emotion shift or confusing labels. We also evaluate ML-ERC with the multi-labels we produced to support our contrastive learning scheme.
@article{kang-cho-2025-beyond,title={Beyond Single Emotion: Multi-label Approach to Conversational Emotion Recognition},author={Kang, Yujin and Cho, Yoon-Sik},journal={Proceedings of the AAAI Conference on Artificial Intelligence},volume={39},number={23},pages={24321--24329},month=feb,year={2025},url={https://ojs.aaai.org/index.php/AAAI/article/view/34609},doi={10.1609/aaai.v39i23.34609},presentation={Oral Presentation}}
2024
EACL
Improving Contrastive Learning in Emotion Recognition in Conversation via Data Augmentation and Decoupled Neutral Emotion
Yujin Kang, and Yoon-Sik Cho
In Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers) , Mar 2024
Emotion recognition in conversation (ERC) has attracted much attention due to its wide applications. While consistent improvement is being made in this area, inevitable challenge comes from the dataset. The ERC dataset exhibits significantly imbalanced emotion distribution. While the utterances with neutral emotion predominate the data, this emotion label is always treated the same as other emotion labels in current approaches. To address the problem caused by the dataset, we propose a supervised contrastive learning specifically oriented for ERC task. We employ a novel data augmentation method emulating the emotion dynamics in a conversation and formulate supervised contrastive learning method tailored for ERC addressing the predominance and the ambiguity of neutral emotion. Experimental results on four benchmark datasets demonstrate the effectiveness of our approach.
@inproceedings{kang-cho-2024-improving,title={Improving Contrastive Learning in Emotion Recognition in Conversation via Data Augmentation and Decoupled Neutral Emotion},author={Kang, Yujin and Cho, Yoon-Sik},booktitle={Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)},month=mar,year={2024},address={St. Julian's, Malta},publisher={Association for Computational Linguistics},url={https://aclanthology.org/2024.eacl-long.134/},doi={10.18653/v1/2024.eacl-long.134},pages={2194--2208},presentation={Oral Presentation}}
2023
IEEE Access
Directed Acyclic Graphs With Prototypical Networks for Few-Shot Emotion Recognition in Conversation
@article{kang-cho-2023-dag,title={Directed Acyclic Graphs With Prototypical Networks for Few-Shot Emotion Recognition in Conversation},author={Kang, Yujin and Cho, Yoon-Sik},journal={IEEE Access},volume={11},pages={117633--117642},year={2023},doi={10.1109/ACCESS.2023.3325893},}