Seongheon Park
Contact. seongheon_park [at] cs [dot] wisc [dot] edu
1210 W Dayton St
Madison, WI 53706
Hello! I’m a second-year PhD student in the Computer Sciences department at the University of Wisconsin-Madison, where I am fortunate to be advised by Prof. Sharon Yixuan Li. Previously, I completed my MS degree at Yonsei University in the Electrical and Electronic Engineering department under the supervision of Prof. Kwanghoon Sohn and Prof. Kibok Lee.
My research interests lie broadly in safe and reliable foundation models, with a current focus on large language models, multi-modal language models, and diffusion models. By “reliable,” I mean developing methods that help users better trust model outputs. Specifically, my work centers on:
- Detecting and mitigating hallucinations and errors
- Uncertainty quantification
- Interpretability
I am also interested in applying these techniques to broader challenges, including the design of more accurate reward and training signals for post-training and self-verification/improvement.
I am actively seeking internship opportunities for Summer 2026. Please feel free to reach out if there is a potential fit!
news
| Sep 18, 2025 | Happy to share that two of our papers have been accepted at NeurIPS 2025 |
|---|---|
| Sep 04, 2025 | Our paper HalluEntity: Benchmarking and Understanding Entity-Level Hallucination Detection is accepted by TMLR |
| Jul 02, 2025 | Our paper about LVLM object hallucnation detection is accepted by ICML 2025 Workshop on Reliable and Responsible Foundation Models |
| May 01, 2025 | Two papers are accepted by ICML 2025 |
| Apr 02, 2025 | Two papers are accepted by ICLR 2025 Workshop: Quantify Uncertainty and Hallucination in Foundation Models |
publications
- Preprint
Understanding Language Prior of LVLMs by Contrasting Chain-of-EmbeddingarXiv:2509.23050, Sep 2025 - Preprint
Shaking to Reveal: Perturbation-Based Detection of LLM HallucinationsarXiv:2506.02696, Jun 2025 - NeurIPS
GLSim: Detecting Object Hallucinations in LVLMs via Global-Local SimilarityConference on Neural Information Processing Systems, Jul 2025 - NeurIPS
GeoRanker: Distance-Aware Ranking for Worldwide Image GeolocalizationConference on Neural Information Processing Systems, May 2025 - ICML
Steer LLM Latents for Hallucination DetectionInternational Conference on Machine Learning, May 2025 - ICML
Position: Challenges and Future Directions of Data-Centric AI AlignmentInternational Conference on Machine Learning, May 2025 - TMLR
HalluEntity: Benchmarking and Understanding Entity-Level Hallucination DetectionTransactions on Machine Learning Research, Mar 2025 - CVPR Workshop
Rethinking Open-World Semi-Supervised Learning: Distribution Mismatch and Inductive InferenceCVPR Workshop on Computer Vision in the Wild, Jun 2024 - ICCV
Hierarchical visual primitive experts for compositional zero-shot learningIEEE/CVF International Conference on Computer Vision, Oct 2023 - CVPR
Partmix: Regularization strategy to learn part discovery for visible-infrared person re-identificationIEEE/CVF Conference on Computer Vision and Pattern Recognition, Jun 2023 - WACV
Normality guided multiple instance learning for weakly supervised video anomaly detectionIEEE/CVF Winter Conference on Applications of Computer Vision, Jan 2023 - WACV
Language-free training for zero-shot video groundingIEEE/CVF Winter Conference on Applications of Computer Vision, Jan 2023