Story
arxiv_agent_systems_research · Nov 9, 2026 · paper
Source brief
Which Successes Make Good References? Action-Sequence Alignment for Multi-Turn On-Policy Distillation
arxiv.orgNov 9, 2026
original source linked
In brief
On-policy distillation (OPD) reduces train-inference distribution mismatch by supervising a student on its own trajectories. For multi-turn tool-use agents, conditioning the teacher on a validated same-task successful...
Feed lens
agenteval