Training-Inference Mismatch In LLM KD(2) View Slides slides #LLM #KD Training-Inference Mismatch In LLM KD(2) https://sophilex.github.io/posts/5489d3fa/ 作者 Sophilex 发布于 2025年10月11日 许可协议 Squeeze, Recover and Relabel: Dataset Condensation at ImageNet Scale From A New Perspective 上一篇 FROM CORRECTION TO MASTERY: REINFORCED DISTILLATION OF LARGE LANGUAGE MODEL AGENTS 下一篇 Please enable JavaScript to view the comments