Seungyoo Lee

MS student, Korea Advanced Institute of Science & Technology

Joined

September 2024

Names

Seungyoo Lee (Preferred)

,

SEUNGYOO LEE

Emails

****@kaist.ac.kr (Confirmed)

Personal Links

Career & Education History

MS student

Korea Advanced Institute of Science & Technology (kaist.ac.kr)

2024 – Present

Advisors, Relations & Conflicts

PhD Advisor

2024 – Present

Expertise

No areas of expertise listed

Publications

Confidence is Not Universal: Task-Dependent Calibration and Emergent Behavior in LLMs
Chaeyun Jang, Moonseok Choi, Yegon Kim, Seungyoo Lee, Juho Lee, Hyungi Lee
- ICML 2026 regular
- Readers: Everyone
From Drift to Coherence: Stabilizing Beliefs in LLMs
SongEun Kim, Seungyoo Lee, Edwin Fong, Hyungi Lee, Juho Lee
- ICML 2026 regular
- Readers: Everyone
OCNR: Stabilizing Self-Play by Mitigating Iteration-Collapse With One-Class Novelty Rewards
Seungyoo Lee, Giung Nam, Hyungi Lee, Juho Lee
- ICML 2026 regular
- Readers: Everyone
Verbalized Confidence Triggers Self-Verification: Emergent Behavior Without Explicit Reasoning Supervision
Chaeyun Jang, Moonseok Choi, Seungyoo Lee, Yegon Kim, Hyungi Lee, Juho Lee
- ICLR 2026 Conference Desk Rejected Submission
- Readers: Everyone
Parallel Test-Time Scaling with Multi-Sequence Verifiers
Yegon Kim, Seungyoo Lee, Chaeyun Jang, Hyungi Lee, Juho Lee
- Submitted to ICLR 2026
- Readers: Everyone
PANGEA: Projection-Based Augmentation with Non-Relevant General Data for Enhanced Domain Adaptation in LLMs
Seungyoo Lee, Giung Nam, Moonseok Choi, Hyungi Lee, Juho Lee
- NeurIPS 2025 poster
- Readers: Everyone
Variational Bayesian Pseudo-Coreset
Hyungi Lee, Seungyoo Lee, Juho Lee
- ICLR 2025 Poster
- Readers: Everyone

Co-Authors