HarnessBandit: Joint Learnability-Transferability Scheduling for Multi-Harness Agentic Reinforcement Learning
This work develops HarnessBandit, an online scheduler that selects one harness per optimizer step, and improves over mixed-batch multi-harness training on both benchmarks, while training diagnostics indicate that learnability and transferability provide distinct, evolving signals.
Hong-Liang Wei, Xiao-Bing Tu, Yinggui Wang et al.
· 1 citation
· ⚡1