Large language model optimization is an active research area, spanning quantization of model weights, early-exit methods for skipping layers, and speculative…
机构:Stanford
来源:arXiv 2609.18005 | AI4Papers 论文推荐平台