斯坦福团队搞了个新招：不用梯度下降，直接把事实写进Transformer的MLP里，论文被COLM 2026收了，想了解怎么不训练就灌知识的可以看看。
斯坦福团队提出一种无需梯度下降即可将事实知识写入Transformer MLP的封闭形式方法。该方法可直接将事实编码进Transformer块，无需训练。相关论文已被COLM 2026接收。研究者包括Jerry Liu、Garctrob、Ronny Junkins等。
This amazing team shows how to build knowledge directly into Transformer blocks **without gradient d...
This amazing team shows how to build knowledge directly into Transformer blocks **without gradient descent**! Jerry Liu @jerrywliu MLPs store facts in language models. Can we write them into Transformers without training? New work w/ amazing team @garctrob @ronnygjunkins @EyubogluSabri , Atri Rudra & @HazyResearch gives a ✨closed-form✨ recipe for fact-storing, Transformer-ready MLPs. Accepted at COLM 2026! 🔗 View Quoted Tweet 💬 1 🔄 1 ❤️ 2 👀 1169 📊 1 ⚡