The case for intelligence beyond language

超越语言的智能：论据

社区

与Google DeepMind研究副总裁Raia Hadsell在2026年RAAIS上的对话

Air Street Press

和

Nathan Benaich

2026年7月21日

文章配音

0:00

-6:35

您的浏览器不支持音频播放。请升级浏览器。

Raia Hadsell在本届RAAIS上指出，Google DeepMind在过去20年最重要的发现是：在足够复杂的系统数据上训练的大型模型，能够学习到该系统的底层规律。语言是这样一个系统，天气、生物学、合成世界和物理运动是其他系统，而她感到沮丧的是，该领域始终将语言视为唯一目标。

Hadsell曾在2017年首届研究与应用人工智能峰会上发表演讲，并在第十届峰会上再次担任开幕演讲。如今她是DeepMind的研究副总裁，领导着一个由约1000名研究人员和工程师组成的前沿AI团队，负责约100个项目。她最初并未从事机器学习领域，本科主修哲学。她的演讲部分是对当前趋势的纠偏。

反对逐词生成的论点

在她发表演讲不到48小时前，DeepMind发布了DiffusionGemma，这是Gemma 4系列下的一个新开源模型，采用Apache 2.0许可证。这是一个260亿参数的专家混合模型，其文本生成方式不同于大多数语言模型。它借鉴了图像和视频生成的方案：先取样，再添加噪声，最后学习去噪。应用于文本时，这意味着一次性生成整个文本块，而不是逐词生成。

这种机制带来了两个优势。推理速度更快，因为模型不需要从左到右逐词处理。同时，由于能一次性生成并修改整个文本块，模型可以双向推理并自我修正。Hadsell展示了它解决某个问题的过程：在第一次去噪步骤中它得出错误答案，第二次重新考虑后修正，到第三次整个文本块变得一致且正确。而自回归模型在锁定每个词后无法回溯，因此无法做到这一点。她说，这正是为什么从左到右的模型在结构上不适合解决数独问题——数独需要同时看到整个网格并进行全局推理。扩散模型还带来了适应性计算的副作用：简单问题只需几步就能收敛，复杂问题则需要更多步骤。实际应用中，它的速度足够快，可以实时与用户保持同步，快速编写、运行和测试代码。不过，DiffusionGemma目前只是一个实验性发布，而非对标准自回归Gemma 4模型的替代，文本扩散本身也存在权衡。DeepMind随模型一同发布了微调代码。

The line runs from Genie-1, trained in 2023 on 2D platformer games and published in 2024, to Genie-2, which moved to latent diffusion and produced interactive 3D environments that were still synthetic and not real-time, to Genie-3, which is high-resolution, works across domains, and runs in real time for several minutes while staying consistent. You feed it a text prompt and get back a world you can move through. Prompt “a country lane in Kent, England,” add rain, and the model fills in the person walking, the puddles, and a pair of yellow boots she is fairly sure she does not own. You can change the world mid-stream by typing a new prompt into it.

Last month the team grounded Genie in Street View. They take one of Street View’s static panoramas, turn it into a short video, prompt in changes such as winter weather or a character, and feed that in as the opening frames of a Genie world, enough to walk across the Golden Gate Bridge and then dive underwater while the model keeps the bridge’s geometry intact. Hadsell is most excited about education: being present at a historical event instead of reading about it, or following a red blood cell through the circulatory system, with the model tied to a textbook chapter so that, in her words, “no dragons will come in if we don’t want them to.”

Robotics is still the hard part

Hadsell has worked on robots for more than a decade and is blunt that the field is hard, an interplay of hardware, software, and understanding, with the goal of general-purpose machines that can drop into any task a human can do. DeepMind splits the problem the way a body does, into a brain and a spine. The spine handles the fast, fine motor control that should not have to route through the brain; the brain handles understanding and planning. Two models carry this. Gemini Robotics-ER 1.6 is a Gemini model tuned for spatial understanding and tool use, able to pick individual objects out of cluttered scenes. A separate vision-language-action model takes the visual input and the reasoner’s instructions and outputs motor control. Working with Boston Dynamics, they found tasks that needed more than spatial sense: getting Spot to read a small gauge required tool calls to magnify the dial and reasoning to interpret it.

The frontier, she said, is efficient learning. The ways to collect robot data run from expensive, narrow teleoperation, through physics simulators like MuJoCo, to the prize of learning directly from internet video. World models now sit in that chain. They can generate novel interaction data at high enough fidelity to train the robot model, and Hadsell said you cannot tell the real robot logs from the simulated ones. Pool that with cross-embodiment learning, where data collected on one robot transfers to others, and the performance gain comes “for free.” This is where the talk’s argument closes on itself: the same foundation-model recipe, pointed at worlds, becomes the engine that trains the robots.

Stop staring at language

她的结语回到了最初表达的沮丧情绪：几乎所有领域的关注点都集中在语言上，而她认为这是资源浪费。今天语言模型所依赖的相同技术路径，完全可以应用于更丰富的系统：世界、机器人、生物、气候，以及地球生物圈在行星尺度上的多样性。相关数据已经存在；相应的模型才刚刚开始构建。按照她的观点，未来十年的突破将围绕这些系统展开，与扩大聊天机器人的规模关系不大。