Modelpedia
Work in progress
About
Findings
Models
Concepts
Methods
Datasets
Sources
Light
Dark
Two Effects, One Trigger: On the Modality Gap, Object Bias, and Information Imbalance in Contrastive Vision-Language Models
2025-01-22
· ICLR 2025 Oral ·
anchor
Findings
IC-038
Few embedding dimensions drive the modality gap in CLIP and SigLIP
IC-039
Object bias in CLIP and SigLIP is not correlated with performance on attribute tasks
IC-040
Information imbalance triggers both the modality gap and object bias in contrastive VLMs
IC-041
CLIP and SigLIP use the modality gap to control logit entropy