文本扩散模型的研究者终于有了一个强大的开放基线——DiffusionGemma 同时生成整段文本并支持实时纠错，做生成式 AI 研究的团队值得下载权重试试。
Google DeepMind 推出 DiffusionGemma，一种基于扩散的新型开放模型，可在专用 GPU 上实现最高 4 倍的输出加速。与传统逐词预测不同，它同时生成整段文本，并能在生成过程中自我纠错和实时格式化复杂 Markdown。该模型权重已在 Hugging Face 上开放。这对于文本扩散领域的研究者来说是一个重要进展，因为该方向目前仍有大量未探索的研究问题。
This is awesome! I am spending a lot of time on diffusion LLMs these days, so this is perfect timin...
This is awesome! I am spending a lot of time on diffusion LLMs these days, so this is perfect timing. I feel like there are so many underexplored research questions around text diffusion. Weight available in HF. Google DeepMind @GoogleDeepMind DiffusionGemma is our new experimental open model with up to 4x faster output on dedicated GPUs. Instead of predicting word-by-word, it generates entire blocks of text simultaneously. This lets the model self-correct and format complex markdown in real time. 🔗 View Quoted Tweet 💬 3 🔄 0 ❤️ 10 👀 835 📊 3 ⚡