Guided by LLM, Grounded in Targeted Vision: Dynamic Chain-of-Thought with Retrieval-Augmented In-Context Learning for Knowledge-Based Visual Question Answering

Yuling Yang, Weizhuo Chen, Cong Cao 0001, Fangfang Yuan, Shuai Zeng, Yanbing Liu 0007. Guided by LLM, Grounded in Targeted Vision: Dynamic Chain-of-Thought with Retrieval-Augmented In-Context Learning for Knowledge-Based Visual Question Answering. In De-Shuang Huang, Chuanlei Zhang, Wei Chen 0036, Bo Li 0002, Qinhu Zhang, Wenzheng Bao, Yijie Pan, Prashan Premaratne, editors, Advanced Intelligent Computing Technology and Applications - 22nd International Conference on Intelligent Computing, ICIC 2026, Toronto, ON, Canada, July 22-26, 2026, Proceedings, Part XIX. Volume 16660 of Lecture Notes in Computer Science, pages 493-504, Springer, 2026. [doi]