Conference
Open access
STARK: Strategic Advantage Reinforcement with Confidence-Driven Exploration for Long-Horizon Agentic Learning
Jin-Yang Wu
Xi-Nan Chang
Yu-Hao Shen
Zhengxi Lu
Shuo Yang
Mingkuan Feng
Jiang-Yan Yi
Shuai Zhang
Zheng Lian
Zheng-Qi Wen
Jian-Hua Tao
2026 · Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing · 0 citations
· 31 references
Abstract
,