-
Self-Augmented Visual Contrastive Decoding
Paper • 2510.13315 • Published -
VidHalluc: Evaluating Temporal Hallucinations in Multimodal Large Language Models for Video Understanding
Paper • 2412.03735 • Published -
Concrete Jungle: Towards Concreteness Paved Contrastive Negative Mining for Compositional Understanding
Paper • 2604.13313 • Published • 12
Eun Woo Im
eunwooim
·
AI & ML interests
multimodal learning
Organizations
None yet
Multimodal-Learning
-
Self-Augmented Visual Contrastive Decoding
Paper • 2510.13315 • Published -
VidHalluc: Evaluating Temporal Hallucinations in Multimodal Large Language Models for Video Understanding
Paper • 2412.03735 • Published -
Concrete Jungle: Towards Concreteness Paved Contrastive Negative Mining for Compositional Understanding
Paper • 2604.13313 • Published • 12
Visual-Restorations