SOTAVerified

Visual Question Answering (VQA)

Visual Question Answering (VQA) is a task in computer vision that involves answering questions about an image. The goal of VQA is to teach machines to understand the content of an image and answer questions about it in natural language.

Image Source: visualqa.org

Papers

Showing 20512100 of 2167 papers

TitleStatusHype
Transfer Learning via Unsupervised Task Discovery for Visual Question AnsweringCode0
What's Different between Visual Question Answering for Machine "Understanding" Versus for Accessibility?Code0
Convincing Rationales for Visual Question Answering ReasoningCode0
Transformer Module Networks for Systematic Generalization in Visual Question AnsweringCode0
Robust Explanations for Visual Question AnsweringCode0
HAIBU-ReMUD: Reasoning Multimodal Ultrasound Dataset and Model Bridging to General Specific DomainsCode0
Attribute Diversity Determines the Systematicity Gap in VQACode0
Visual Contexts Clarify Ambiguous Expressions: A Benchmark DatasetCode0
Visual Coreference Resolution in Visual Dialog using Neural Module NetworksCode0
Transparency by Design: Closing the Gap Between Performance and Interpretability in Visual ReasoningCode0
Hadamard Product for Low-rank Bilinear PoolingCode0
Routing Networks and the Challenges of Modular and Compositional ComputationCode0
RSAdapter: Adapting Multimodal Models for Remote Sensing Visual Question AnsweringCode0
Guiding Vision-Language Model Selection for Visual Question-Answering Across Tasks, Domains, and Knowledge TypesCode0
Contrastive Visual-Linguistic PretrainingCode0
Evaluating Point Cloud from Moving Camera Videos: A No-Reference MetricCode0
Grounding Answers for Visual Questions Asked by Visually Impaired PeopleCode0
RUBi: Reducing Unimodal Biases for Visual Question AnsweringCode0
RUBi: Reducing Unimodal Biases in Visual Question AnsweringCode0
Grad-CAM: Why did you say that?Code0
GradBias: Unveiling Word Influence on Bias in Text-to-Image Generative ModelsCode0
RVTBench: A Benchmark for Visual Reasoning TasksCode0
Attention on Attention: Architectures for Visual Question Answering (VQA)Code0
Ask Your Neurons: A Deep Learning Approach to Visual Question AnsweringCode0
Generalizing Visual Question Answering from Synthetic to Human-Written Questions via a Chain of QA with a Large Language ModelCode0
General Greedy De-bias LearningCode0
What's in a Question: Using Visual Questions as a Form of SupervisionCode0
A Neuro-Symbolic ASP Pipeline for Visual Question AnsweringCode0
GAMIVAL: Video Quality Prediction on Mobile Cloud Gaming ContentCode0
An Efficient Modern Baseline for FloodNet VQACode0
Black-box Model Ensembling for Textual and Visual Question Answering via Information FusionCode0
Game of Sketches: Deep Recurrent Models of Pictionary-style Word GuessingCode0
Ask, Attend and Answer: Exploring Question-Guided Spatial Attention for Visual Question AnsweringCode0
TUBench: Benchmarking Large Vision-Language Models on Trustworthiness with Unanswerable QuestionsCode0
Tutorial on Answering Questions about Images with Deep LearningCode0
Scene Graph Prediction with Limited LabelsCode0
Continual VQA for Disaster Response SystemsCode0
What value do explicit high level concepts have in vision to language problems?Code0
FVQ: A Large-Scale Dataset and A LMM-based Method for Face Video Quality AssessmentCode0
Two-Level Approach for No-Reference Consumer Video Quality AssessmentCode0
Zero-shot Visual Question Answering with Language Model FeedbackCode0
Fully Authentic Visual Question Answering Dataset from Online CommunitiesCode0
Analyzing the Behavior of Visual Question Answering ModelsCode0
Adapting Visual Question Answering Models for Enhancing Multimodal Community Q&A PlatformsCode0
From Images to Textual Prompts: Zero-shot VQA with Frozen Large Language ModelsCode0
A simple neural network module for relational reasoningCode0
Visually Grounded VQA by Lattice-based RetrievalCode0
VQA4CIR: Boosting Composed Image Retrieval with Visual Question AnsweringCode0
UGC Quality Assessment: Exploring the Impact of Saliency in Deep Feature-Based Quality AssessmentCode0
FRAMES-VQA: Benchmarking Fine-Tuning Robustness across Multi-Modal Shifts in Visual Question AnsweringCode0
Show:102550
← PrevPage 42 of 44Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1humanAccuracy89.3Unverified
2DREAM+Unicoder-VL (MSRA)Accuracy76.04Unverified
3TRRNet (Ensemble)Accuracy74.03Unverified
4MIL-nbgaoAccuracy73.81Unverified
5Kakao BrainAccuracy73.33Unverified
6Coarse-to-Fine Reasoning, Single ModelAccuracy72.14Unverified
7270Accuracy70.23Unverified
8NSM ensemble (updated)Accuracy67.55Unverified
9VinVL-DPTAccuracy64.92Unverified
10VinVL+LAccuracy64.85Unverified
#ModelMetricClaimedVerifiedStatus
1PaLIAccuracy84.3Unverified
2BEiT-3Accuracy84.19Unverified
3VLMoAccuracy82.78Unverified
4ONE-PEACEAccuracy82.6Unverified
5mPLUG (Huge)Accuracy82.43Unverified
6CuMo-7BAccuracy82.2Unverified
7X2-VLM (large)Accuracy81.9Unverified
8MMUAccuracy81.26Unverified
9InternVL-CAccuracy81.2Unverified
10LyricsAccuracy81.2Unverified
#ModelMetricClaimedVerifiedStatus
1BEiT-3overall84.03Unverified
2mPLUG-Hugeoverall83.62Unverified
3ONE-PEACEoverall82.52Unverified
4X2-VLM (large)overall81.8Unverified
5VLMooverall81.3Unverified
6SimVLMoverall80.34Unverified
7X2-VLM (base)overall80.2Unverified
8VASToverall80.19Unverified
9VALORoverall78.62Unverified
10Prompt Tuningoverall78.53Unverified