arXiv cs.CL
· Papers
Student-Centered Distillation Narrows the Agentic Gap Between Small and Large LLMs
arXiv:2509.14257v3 Announce Type: replace Abstract: Large Language Model agents achieve strong performance on multi-step reasoning and tool-use tasks, but their impressive capabilities typically rely on extremely large backbones. Existing distillation approaches train smaller students to imitate full teacher trajectori