Ablations show that selective intervention outperforms passive bank exposure, always-on injection, advisor-only guidance, advisor-only guidance, and general retrieval, and general retrieval and that selective intervention outperforms passive bank exposure, always-on injection, advisor-only guidance, and general retrieval.
Yi-Fan Wu, Lizhu Zhang, Yuhang Zhou et al.· 1 citation
A novel alignment framework, Reinforcement Learning for Selection Reward (RLSR), is proposed, which targets the area under the risk-coverage curve (AURC) -- a popular SP performance metric -- as its alignment objective and achieves substantially better risk-coverage trade-off compared to multiple alignment baselines on both in-domain and out-of-domain tasks.
Gaoxiang Luo, Yi-Fan Wu, Sinian Zhang et al.· 0 citations
EvoHarness-RL is introduced, which exposes Belief, Progress, and Experience (BPE) as policy-facing harness state and reveals two key dynamics: harness annealing, where training internalizes recurring harness-use patterns into the model policy and shifts the agent from frequent harness calls toward selective external-state access, and harness evolution, where progress updates and experience consolidation refine the harness into a compact, task-adaptive state substrate.
Xuying Ning, Dongqi Fu, Tianxin Wei et al.· 0 citations