SING: Synthetic Intention Graph for Scalable Active Tool Discovery in LLM Agents
Published in arXiv:2606.16591, 2026
SING connects user intentions, tool capabilities, and tool collaboration patterns in a graph to help language model agents discover relevant tools as tasks evolve.
Recommended citation: Qiao Xiao, Haochen Shi, Yisen Gao, Wenbin Hu, Huihao Jing, Tianshi Zheng, Baixuan Xu, Ziheng Zhang, Weiqi Wang, Haoran Li, Jiaxin Bai, and Yangqiu Song. 2026. SING: Synthetic Intention Graph for Scalable Active Tool Discovery in LLM Agents. arXiv:2606.16591.
Download Paper