HyperAIHyperAI초신경
홈뉴스연구 논문튜토리얼데이터셋백과사전SOTALLM 모델GPU 랭킹컨퍼런스
전체 검색
소개
HyperAIHyperAI초신경

Visual Entailment On Snli Ve Val

평가 지표

Accuracy

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름
Accuracy
Paper TitleRepository
CLIP-ViL80.20How Much Can CLIP Benefit Vision-and-Language Tasks?
EVE-ROI*70.81Visual Entailment: A Novel Task for Fine-Grained Image Understanding
OFA91.0OFA: Unifying Architectures, Tasks, and Modalities Through a Simple Sequence-to-Sequence Learning Framework
SimVLM86.21SimVLM: Simple Visual Language Model Pretraining with Weak Supervision
UNITER78.98UNITER: UNiversal Image-TExt Representation Learning
Prompt Tuning90.04Prompt Tuning for Generative Multimodal Pretrained Models
VILLA-LARGE80.18Large-Scale Adversarial Training for Vision-and-Language Representation Learning
CoCa87.0CoCa: Contrastive Captioners are Image-Text Foundation Models
SOHO85.00Seeing Out of tHe bOx: End-to-End Pre-training for Vision-Language Representation Learning
0 of 9 row(s) selected.