HyperAIHyperAI超神経
ホームニュース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 Reasoning On Nlvr2 Dev

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
XFM (base)87.6Toward Building General Foundation Models for Language, Vision, and Vision-Language Understanding Tasks
VLMo85.64VLMo: Unified Vision-Language Pre-Training with Mixture-of-Modality-Experts
VisualBERT66.7VisualBERT: A Simple and Performant Baseline for Vision and Language
VK-OOD83.9Differentiable Outlier Detection Enable Robust Deep Multimodal Analysis
CoCa86.1CoCa: Contrastive Captioners are Image-Text Foundation Models
X-VLM (base)84.41Multi-Grained Vision Language Pre-Training: Aligning Texts with Visual Concepts
X2-VLM (large)88.7X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks
SOHO76.37Seeing Out of tHe bOx: End-to-End Pre-training for Vision-Language Representation Learning
ALBEF (14M)83.14Align before Fuse: Vision and Language Representation Learning with Momentum Distillation
VK-OOD84.6Implicit Differentiable Outlier Detection Enable Robust Deep Multimodal Analysis
ViLT-B/3275.7ViLT: Vision-and-Language Transformer Without Convolution or Region Supervision
SimVLM84.53SimVLM: Simple Visual Language Model Pretraining with Weak Supervision
BEiT-391.51Image as a Foreign Language: BEiT Pretraining for All Vision and Vision-Language Tasks
X2-VLM (base)86.2X$^2$-VLM: All-In-One Pre-trained Model For Vision-Language Tasks
LXMERT (Pre-train + scratch)74.9LXMERT: Learning Cross-Modality Encoder Representations from Transformers
0 of 15 row(s) selected.