Scholar
Sanghwan Kim
Google Scholar ID: XkqHJfYAAAAJ
Technical University of Munich, Helmholtz Munich
machine learning and deep learning
Follow
Google Scholar
↗
Citations & Impact
All-time
Citations
139
H-index
6
i10-index
3
Publications
8
Co-authors
0
Publications
7 items
VisLens: Single-Pass Interpretable Visual Search for Multimodal LLMs
2026
Cited
0
From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA
2026
Cited
0
When Sinks Help or Hurt: Unified Framework for Attention Sink in Large Vision-Language Models
2026
Cited
0
FINER: MLLMs Hallucinate under Fine-grained Negative Queries
2026
Cited
0
From Drop-off to Recovery: A Mechanistic Analysis of Segmentation in MLLMs
2026
Cited
0
Training-free Uncertainty Guidance for Complex Visual Tasks with MLLMs
2025
Cited
0
COSMOS: Cross-Modality Self-Distillation for Vision Language Pre-training
arXiv.org · 2024
Cited
0
