用更一致的轨迹、更少的解码步数「驯服」掩码扩散语言模型，扩散语言模型的推理性能和效率大幅提升
"扩散大语言模型" 相关文章
Attention Is All You Need for KV Cache in Diffusion LLMs
Attention Is All You Need for KV Cache in Diffusion LLMs
Boundary-Guided Policy Optimization for Memory-efficient RL of Diffusion Large Language Models
SPG: Sandwiched Policy Gradient for Masked Diffusion Language Models
Finish First, Perfect Later: Test-Time Token-Level Cross-Validation for Diffusion Large Language Models
Finish First, Perfect Later: Test-Time Token-Level Cross-Validation for Diffusion Large Language Models
Principled and Tractable RL for Reasoning with Diffusion Language Models
免训练加速61倍！陈怡然团队新作DPad：仅关注「彩票token」
免训练加速61倍！陈怡然团队新作DPad：仅关注「彩票token」
免训练加速61倍！陈怡然团队新作DPad：仅关注「彩票token」
扩散大语言模型也能飞？DPad免训练加速61倍，全局规划照样稳
扩散LLM推理新范式：打破生成长度限制，实现动态自适应调节
