Hugging Face Blog·· 2026-08-10AI 评分55
Multiverse Computing 提出低成本知识蒸馏方法，长上下文训练可单卡运行
Making Knowledge Distillation Cheap Enough to Run at Scale
AI 导读
Multiverse Computing 发布论文《Efficient Knowledge Distillation for LLMs: Offline Top-K Logits and a Fused Chunked KL Loss》。
来源：Hugging Face Blog · huggingface.co