Scholar
Seongheon Park
Google Scholar ID: 8FFbyJ0AAAAJ
University of Wisconsin-Madison
Citations & Impact
Citations
267
H-index
6
i10-index
6
Publications
12
Co-authors
19
list available
Publications
Resume (English only)
Academic Achievements
- Two papers accepted at NeurIPS 2025: 'GLSim: Detecting Object Hallucinations in LVLMs via Global-Local Similarity' and 'GeoRanker: Distance-Aware Ranking for Worldwide Image Geolocalization'
- Two papers accepted at ICML 2025: 'Steer LLM Latents for Hallucination Detection' and 'Position: Challenges and Future Directions of Data-Centric AI Alignment'
- Paper accepted at TMLR 2025: 'HalluEntity: Benchmarking and Understanding Entity-Level Hallucination Detection'
- Two papers accepted at ICLR 2025 Workshop: Quantify Uncertainty and Hallucination in Foundation Models
- One paper accepted at ICML 2025 Workshop on Reliable and Responsible Foundation Models
- Paper accepted at CVPR Workshop 2024: 'Rethinking Open-World Semi-Supervised Learning'
- Paper accepted at ICCV 2023: 'Hierarchical visual primitive experts for compositional zero-shot learning'
- Paper accepted at CVPR 2023: 'Partmix: Regularization strategy to learn part discovery for visible-infrared person re-identification'
- Paper accepted at WACV 2023: 'Normality guided multiple instance learning for weakly supervised video anomaly detection'
Background
- Second-year PhD student in the Computer Sciences department at the University of Wisconsin-Madison
- Research interests broadly lie in safe and reliable foundation models
- Current focus includes large language models, multi-modal language models, and diffusion models
- Aims to enhance trustworthiness of model outputs through hallucination/error detection and mitigation, uncertainty quantification, and interpretability
- Interested in applying these techniques to broader challenges such as designing more accurate reward and training signals for post-training and self-verification/improvement
