Boosting Knowledge-based Visual Question Answering with Structured Context Reasoning
基于结构化上下文推理提升基于知识的视觉问答性能
机构 * School of Computer Science, Guangdong University of Technology(广东工业大学计算机学院) ; School of Artificial Intelligence, Shenzhen University(深圳大学人工智能学院) ; Graduate School of Advanced Science and Engineering, Hiroshima University(广岛大学先进科学与工程研究生院)
专题命中 逻辑推理 :reasoning(title,abstract)
AI总结 本文提出SCoRe框架,通过上下文获取、选择、压缩三阶段处理多模态知识,在OK-VQA和A-OKVQA基准上性能优于现有最优方法,提升了基于知识的视觉问答效果。
Comments Accepted by ICME 2026. Source code is available at this https URL (https://github.com/WISLab-GDUT/SCoRe)