Targeted Error Correction in Knowledge Distillation: Small Language Models Surpass GPT

Open in new window