最近的文章
面向大语言模型的门控注意力机制：非线性、稀疏性和Attention-Sink-Free
面向大语言模型的门控注意力机制：非线性、稀疏性和 Attention-Sink-Free #
Saber：一种针对扩散语言模型的自适应加速与回溯增强的高效采样方法
Saber：一种针对扩散语言模型的自适应加速与回溯增强的高效采样方法 # Saber: An Efficient Sampling with Adaptive Acceleration and Backtracking Enhanced Remasking for Diffusion Language Model
R-STITCH：用于高效推理的动态轨迹拼接
R-STITCH：用于高效推理的动态轨迹拼接 # R-STITCH: DYNAMIC TRAJECTORY STITCHING FOR EFFICIENT REASONING
真-Self-Spec-DLM
真-Self-Spec-DLM # SELF SPECULATIVE DECODING FOR DIFFUSION LARGE LANGUAGE MODELS
SlowFast采样加速DLM
SlowFast 采样加速 DLM # ACCELERATING DIFFUSION LARGE LANGUAGE MODELS WITH SLOWFAST SAMPLING: THE THREE GOLDEN PRINCIPLES
