A-OKVQA

A-OKVQA is a dataset that contains visual question-answering tasks designed to evaluate the compositional visual reasoning abilities of vision-language models.

Papers using A-OKVQA (16)

Self-Questioning Vision-Language Models: Reinforcement Learning for Compositional Visual Reasoning2026Vision Verification Enhanced Fusion of VLMs for Efficient Visual Reasoning2026From Hindsight to Foresight: Self-Encouraged Hindsight Distillation for Knowledge-based Visual Question Answering2025Context-Aware Multi-Turn Visual-Textual Reasoning in LVLMs via Dynamic Memory and Adaptive Visual Guidance2025MV-CoRe: Multimodal Visual-Conceptual Reasoning for Complex Visual Question Answering2025Coherent Multimodal Reasoning with Iterative Self-Evaluation for Vision-Language Models2025Hierarchical Contextual Grounding LVLM: Enhancing Fine-Grained Visual-Language Understanding with Robust Grounding2025See the Forest and the Trees: A Synergistic Reasoning Framework for Knowledge-Based Visual Question Answering2025NLKI: A Lightweight Natural Language Knowledge Integration Framework For Improving Small Vlms In Commonsense VQA Tasks2025Believing Without Seeing: Quality Scores For Contextualizing Vision-language Model Explanations2025Cross Domain Evaluation Of Multimodal Chain-of-thought Reasoning Of Different Datasets Into The Amazon Cot Framework2025Retrieval-Based Interleaved Visual Chain-of-Thought in Real-World Driving Scenarios2025Knowledge Condensation and Reasoning for Knowledge-based VQA2024 · 2 citesII-MMR: Identifying and Improving Multi-modal Multi-hop Reasoning in Visual Question Answering2024 · 1 citesA Simple Baseline for Knowledge-Based Visual Question Answering2023Retrieval Meets Reasoning: Even High-school Textbook Knowledge Benefits Multimodal Reasoning2024