HyperAIHyperAI超神経
ホームニュース最新論文チュートリアルデータセット百科事典SOTALLMモデルGPU ランキング学会
検索
サイトについて
HyperAIHyperAI超神経

Visual Question Answering On Vcr Q A 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 TitleRepository
VL-BERTLARGE75.8VL-BERT: Pre-training of Generic Visual-Linguistic Representations
MAD (Single Model, Formerly CLIP-TD)79.6Multimodal Adaptive Distillation for Leveraging Unimodal Encoders for Vision-Language Tasks-
UNITER (Large)77.3UNITER: UNiversal Image-TExt Representation Learning
GPT4RoI89.4GPT4RoI: Instruction Tuning Large Language Model on Region-of-Interest
VisualBERT71.6VisualBERT: A Simple and Performant Baseline for Vision and Language
ERNIE-ViL-large(ensemble of 15 models)81.6ERNIE-ViL: Knowledge Enhanced Vision-Language Representations Through Scene Graph-
UNITER-large (10 ensemble)79.8UNITER: UNiversal Image-TExt Representation Learning
OFA-X71.2Harnessing the Power of Multi-Task Pretraining for Ground-Truth Level Natural Language Explanations
OFA-X-MT62Harnessing the Power of Multi-Task Pretraining for Ground-Truth Level Natural Language Explanations
VL-T575.3Unifying Vision-and-Language Tasks via Text Generation
KVL-BERTLARGE76.4KVL-BERT: Knowledge Enhanced Visual-and-Linguistic BERT for Visual Commonsense Reasoning-
0 of 11 row(s) selected.
Visual Question Answering On Vcr Q A Test | SOTA | HyperAI超神経