HyperAIHyperAI
HomeNewsLatest PapersTutorialsDatasetsWikiSOTALLM ModelsGPU LeaderboardEvents
Search
About
HyperAIHyperAI

Visual Question Answering On Gqa Test Dev

Metrics

Accuracy

Results

Performance results of various models on this benchmark

Model Name
Accuracy
Paper TitleRepository
BLIP-2 ViT-G OPT 2.7B (zero-shot)34.6BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models-
BLIP-2 ViT-G FlanT5 XXL (zero-shot)44.7BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models-
PNP-VQA41.9Plug-and-Play VQA: Zero-shot VQA by Conjoining Large Pretrained Models with Zero Training-
PaLI-X-VPD67.3Visual Program Distillation: Distilling Tools and Programmatic Reasoning into Vision-Language Models-
LXMERT (Pre-train + scratch)60.0LXMERT: Learning Cross-Modality Encoder Representations from Transformers-
BLIP-2 ViT-L FlanT5 XL (zero-shot)44.4BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models-
FewVLM (zero-shot)29.3A Good Prompt Is Worth Millions of Parameters: Low-resource Prompt-based Learning for Vision-Language Models-
BLIP-2 ViT-G OPT 6.7B (zero-shot)36.4BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models-
HYDRA47.9HYDRA: A Hyper Agent for Dynamic Compositional Visual Reasoning-
NSM62.95Learning by Abstraction: The Neural State Machine-
Lyrics62.4Lyrics: Boosting Fine-grained Language-Vision Alignment and Comprehension via Semantic-aware Visual Objects-
BLIP-2 ViT-L OPT 2.7B (zero-shot)33.9BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models-
single-hop + LCGN (ours)55.8Language-Conditioned Graph Networks for Relational Reasoning-
CFR72.1Coarse-to-Fine Reasoning for Visual Question Answering-
BLIP-2 ViT-G FlanT5 XL (zero-shot)44.2BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models-
Video-LaVIT64.4Video-LaVIT: Unified Video-Language Pre-training with Decoupled Visual-Motional Tokenization-
CuMo-7B64.9CuMo: Scaling Multimodal LLM with Co-Upcycled Mixture-of-Experts-
0 of 17 row(s) selected.
Visual Question Answering On Gqa Test Dev | SOTA | HyperAI