AI RESEARCH

LARK: Learnability-Grounded Trajectory Selection for Efficient Reasoning Distillation

arXiv CS.AI

ArXi:2605.30651v1 Announce Type: cross We study trajectory selection for reasoning distillation, where teacher-generated reasoning trajectories are selectively used as supervision for a student model. Existing methods rely on heuristics such as trajectory quality or model confidence, but they often overlook whether a trajectory is learnable by the student. In this paper, we present LARK, a learnability-grounded method for reasoning trajectory selection. LARK selects trajectories that the student can distribution.