谷歌DiffusionGemma：用不到10%预算将Gemma改造成扩散模型
Google's DiffusionGemma proves you don't need to train from scratch to build a text diffusion model
Instead of training a new model from scratch, Google DeepMind retrofitted Gemma 4 into a diffusion model using less than 10 percent of the original training budget. DiffusionGemma generates 256 tokens in parallel instead of one at a time, hitting about 1,500 tokens per second. Quality still trails the original autoregressive model in benchmarks, especially on reasoning tasks.
谷歌DeepMind没有从头训练新模型，而是用不到原始训练预算10%的成本，将Gemma 4改造为扩散模型。DiffusionGemma并行生成256个令牌，而非逐个生成，每秒约处理1500个令牌。在基准测试中，其质量仍不及原始自回归模型，尤其在推理任务上。
The article Google's DiffusionGemma proves you don't need to train from scratch to build a text diffusion model appeared first on The Decoder.
文章《谷歌的DiffusionGemma证明构建文本扩散模型无需从头训练》首次出现在The Decoder上。
更进一步：量化金融体系
看懂新闻只是起点——沿量化金融路径，把它变成能交付的工程能力