1
OpenAI’s Head of Safety Is Leaving the Company
OpenAI安全负责人离职，内部安全挑战随模型训练加速而加剧，引发行业关注
Johannes Heidecke’s departure comes as OpenAI tries to further integrate its research and safety teams.
2
S2D2: Fast Decoding for Diffusion LLMs via Training-Free Self-Speculation
块扩散语言模型新突破：无需训练的自推测解码，兼顾质量与速度，少步数下比标准方法更稳健。
arXiv:2603.25702v2 Announce Type: replace Abstract: Block-diffusion language models offer a promising path toward faster-than-autoregressive generatio…
3
FABSVer: Faster Training and Better Self-Verification for LLM Mathematical Reasoning
一种新型训练框架，在加速大模型数学推理训练的同时显著提升自我验证能力，精准解决LLM推理瓶颈。
arXiv:2605.28389v1 Announce Type: new Abstract: While large language models have made significant progress in mathematical reasoning, they remain unre…
4
Understanding, Accelerating, and Improving MeanFlow Training
从理解到加速，全方位优化MeanFlow训练方法，为流模型训练效率带来显著提升。
arXiv:2511.19065v2 Announce Type: replace-cross Abstract: MeanFlow promises high-quality generative modeling in few steps, by jointly learning instant…