| Paper | Code | Accuracy | ModelName | ReleaseDate |
|---|---|---|---|---|
| Demystifying CLIP Data | ✓ Link | 52.7 | MetaCLIP ViT-bigG/14 | 2023-09-28 |
| Multi-label Cluster Discrimination for Visual Representation Learning | ✓ Link | 39.4 | MLCD ViT-L/14 | 2024-07-24 |
| SmartCLIP: Modular Vision-language Alignment with Identification Guarantees | ✓ Link | 30.4 | SmartCLIP (ViT-L/14) | 2025-07-29 |
| LeVLJEPA: End-to-End Vision-Language Pretraining Without Negatives | ✓ Link | 7.65 | LeVLJEPA ViT-B/16 (zero-shot) | 2026-07-01 |
| FFF: Fixing Flawed Foundations in Contrastive Pre-Training Results in Very Strong Vision-Language Models | 5.3 | FFF-ViT-B/32 (Open70M) (extra data) | 2024-05-16 |