1
General and Efficient Steering of Diffusion Models
面向扩散模型的通用高效引导框架，大幅提升生成控制灵活性与计算效率
arXiv:2602.11395v2 Announce Type: replace Abstract: Steering diffusion models toward conditions unseen during training typically requires either retra…
2
DSpark: Speculative decoding accelerates LLM inference [pdf]
深度揭秘DeepSeek最新研究：投机解码如何大幅加速LLM推理，论文全文在此。
Article URL: https://github.com/deepseek-ai/DeepSpec/blob/main/DSpark_paper.pdf Comments URL: https://news.ycombinator.com/item?id=48696585 Points: 52…
3
GRINQH: Graded Input-based Quantization Hierarchy for Efficient LLM Generation
量化分层按输入分级，显著提升LLM生成效率，值得算法工程师细读。
arXiv:2606.23419v1 Announce Type: cross Abstract: Autoregressive decoding with LLMs is primarily bottlenecked by GPU memory bandwidth, especially in e…
4
DSB: Dynamic Sliding Block Scheduling for Diffusion LLMs
ICML 2026 接收，提出动态滑动块调度策略，让扩散大语言模型生成效率更高。
arXiv:2602.05992v3 Announce Type: replace Abstract: Diffusion large language models (dLLMs) have emerged as a promising alternative for text generatio…
5
Geometric Latent Reasoning Induces Shorter Generations in LLMs
用几何潜在推理突破传统语言链，让LLM生成更短、更便宜且不牺牲准确性，重新定义推理效率。
arXiv:2606.02248v1 Announce Type: new Abstract: Large language models solve complex problems by generating lengthy chains of explicit reasoning tokens…