ReAD: Reinforcement-Guided Capability Distillation for Large Language Models

arXiv:2605.11290v1 Announce Type: cross Abstract: Capability distillation applies knowledge distillation to selected model capabilities, aiming to compress a large language model (LLM) into a smaller one while preserving the abilities needed for a downstream task. However, most existing methods treat capabilities as independent training targets and overlook how improving one capability can reshape the student's broader capability profile, especially when multiple abilities jointly determine task