HyperAI超神経

プラットフォーム

Visual Entailment On Snli Ve Val

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

		Paper Title
OFA	91.0	OFA: Unifying Architectures, Tasks, and Modalities Through a Simple Sequence-to-Sequence Learning Framework
Prompt Tuning	90.04	Prompt Tuning for Generative Multimodal Pretrained Models
CoCa	87.0	CoCa: Contrastive Captioners are Image-Text Foundation Models
SimVLM	86.21	SimVLM: Simple Visual Language Model Pretraining with Weak Supervision
SOHO	85.00	Seeing Out of tHe bOx: End-to-End Pre-training for Vision-Language Representation Learning
CLIP-ViL	80.20	How Much Can CLIP Benefit Vision-and-Language Tasks?
VILLA-LARGE	80.18	Large-Scale Adversarial Training for Vision-and-Language Representation Learning
UNITER	78.98	UNITER: UNiversal Image-TExt Representation Learning
EVE-ROI*	70.81	Visual Entailment: A Novel Task for Fine-Grained Image Understanding

0 of 9 row(s) selected.

Visual Entailment On Snli Ve Val | SOTA | HyperAI超神経