VQA v-2

VQA-2.0 is a benchmark dataset that contains images paired with questions and answers, used to evaluate the performance of models in visual question answering tasks.

Papers using VQA v-2 (20)

Analyzing the Sensitivity of Vision Language Models in Visual Question Answering2025Bilinear Attention Networks2018 · 78 citesUnified Vision-Language Pre-Training for Image Captioning and VQA2019 · 74 citesExploring Human-like Attention Supervision in Visual Question Answering2017 · 45 citesMUREL: Multimodal Relational Reasoning for Visual Question Answering2019 · 24 citesDynamic Fusion with Intra- and Inter- Modality Attention Flow for Visual Question Answering2018 · 17 citesAnswer-Me: Multi-Task Open-Vocabulary Visual Question Answering2022 · 16 citesCompact Trilinear Interaction for Visual Question Answering2019 · 14 citesSurvey of Recent Advances in Visual Question Answering2017 · 9 citesLeveraging Visual Question Answering to Improve Text-to-Image Synthesis2020 · 7 citesDual Recurrent Attention Units for Visual Question Answering2018 · 6 citesImproved Fusion of Visual and Language Representations by Dense Symmetric Co-Attention for Visual Question Answering2018 · 2 citesMixGen: A New Multi-Modal Data Augmentation2022 · 2 citesCompound Tokens: Channel Fusion for Vision-Language Representation Learning2022 · 2 citesAnswer-checking in Context: A Multi-modal FullyAttention Network for Visual Question Answering2020 · 1 citesEnhancing Visual Question Answering through Ranking-Based Hybrid Training and Multimodal Fusion2024 · 1 citesOpen-Ended Visual Question Answering by Multi-Modal Domain Adaptation2019Image Captioning for Effective Use of Language Models in Knowledge-Based Visual Question Answering2021VQA with Cascade of Self- and Co-Attention Blocks2023Cycle-Consistency for Robust Visual Question Answering2019