Visual Entailment On Snli Ve Test
評価指標
Accuracy
評価結果
このベンチマークにおける各モデルのパフォーマンス結果
モデル名 | Accuracy | Paper Title | Repository |
|---|---|---|---|
| UNITER (Large) | 78.98 | UNITER: UNiversal Image-TExt Representation Learning | - |
| EVE-ROI* | 70.47 | Visual Entailment: A Novel Task for Fine-Grained Image Understanding | - |
| MAD (Single Model, Formerly CLIP-TD) | 80.32 | Multimodal Adaptive Distillation for Leveraging Unimodal Encoders for Vision-Language Tasks | - |
| SimVLM | 86.32 | SimVLM: Simple Visual Language Model Pretraining with Weak Supervision | - |
| SOHO | 84.95 | Seeing Out of tHe bOx: End-to-End Pre-training for Vision-Language Representation Learning | - |
| Prompt Tuning | 90.12 | Prompt Tuning for Generative Multimodal Pretrained Models | - |
| CoCa | 87.1 | CoCa: Contrastive Captioners are Image-Text Foundation Models | - |
| OFA | 91.2 | OFA: Unifying Architectures, Tasks, and Modalities Through a Simple Sequence-to-Sequence Learning Framework | - |
0 of 8 row(s) selected.