5 papers
Selective Test-Time Debiasing for CLIP via Reward Gating
Jaeho Han, Jisoo Yang, Hyeondong Woo +3
Vision language models (VLMs) demonstrate strong zero-shot performance, but often perpetuate social stereotypes in person-centric queries, yielding skewed demographic distributions…
Visual Funnel: Resolving Contextual Blindness in Multimodal Large Language Models
Woojun Jung, Jaehoon Go, Mingyu Jeon +2
Multimodal Large Language Models (MLLMs) demonstrate impressive reasoning capabilities, but often fail to perceive fine-grained visual details, limiting their applicability in prec…
See More, Store Less: Memory-Efficient Resolution for Video Moment Retrieval
Mingyu Jeon, Sungjin Han, Jinkwon Hwang +3
Recent advances in Multimodal Large Language Models (MLLMs) have improved image recognition and reasoning, but video-related tasks remain challenging due to memory constraints from…
GranAlign: Granularity-Aware Alignment Framework for Zero-Shot Video Moment Retrieval
Mingyu Jeon, Sunjae Yoon, Jonghee Kim +1
Zero-shot video moment retrieval (ZVMR) is the task of localizing a temporal moment within an untrimmed video using a natural language query without relying on task-specific traini…
Point to Span: Zero-Shot Moment Retrieval for Navigating Unseen Hour-Long Videos
Mingyu Jeon, Jisoo Yang, Sungjin Han +4
Zero-shot Long Video Moment Retrieval (ZLVMR) is the task of identifying temporal segments in hour-long videos using a natural language query without task-specific training. The co…