2025
Fix-CLIP: Dual-Branch Hierarchical Contrastive Learning via Synthetic Captions for Better Understanding of Long Text
ICCV 2025poster
CLIP has shown promising performance across many short-text tasks in a zero-shot manner. However, limited by the input length of the text encoder, CLIP struggles on under-stream tasks with long-text inputs (>77 tokens). To remedy this issue, we propose FIX-CLIP, which includes three novel modules: (…