Papers by Suhyun Lee
Sentimatic: Sentiment-guided Automatic Generation of Preference Datasets for Customer Support Dialogue System (2025.naacl-srw)
Copied to clipboard
| Challenge: | Existing methods for supervised fine-tuning and preference optimization (PO) rely on human-authored evaluation criteria for practical application. |
| Approach: | They propose a method to generate customer preference datasets without human intervention using a publicly available dataset constructed for SFT. |
| Outcome: | The proposed method classifies responses by sentiment, fine-tunes models on them, and applies advanced sampling and evaluation techniques to ensure diversity and quality. |
MHSafeEval: Role-Aware Interaction-Level Evaluation of Mental Health Safety in Large Language Models (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing evaluation frameworks assess isolated responses using coarse-grained taxonomies or static datasets. |
| Approach: | They propose a role-aware mental health safety taxonomy that characterizes clinically significant harm in terms of interactional roles an AI counselor adopts. |
| Outcome: | The proposed framework significantly improves failure-mode coverage and diagnostic granularity. |