PETS: A Principled Framework Towards Optimal Trajectory Allocation for Efficient Test-Time Self-Consistency 事件

PETS: A Principled Framework Towards Optimal Trajectory Allocation for Efficient Test-Time Self-Consistency arXiv:2602.16745v2 Announce Type: replace-cross Abstract: Test-time scaling can improve model performance by aggregating stochastic reasoning trajectories. However, achieving sample-efficient test-time self-consistency under a limited budget remains an open challenge. We introduce PETS (Principled and Efficient Test-TimeSelf-Consistency), which initiates a principled study of trajectory a