Industry

François Chollet:当前 Transformer 技术栈并非最高效的学习方案

Why we picked it

Keras 之父 François Chollet 直接放话:Transformer 加梯度下降这套组合,效率连前 50% 都排不进。做架构研究或长期押注某个技术路线的人,值得看看他对下一代的判断。

François Chollet 在 X 上发文称,学习算法的可能性空间极为广阔,覆盖长期可塑性(预训练)与即时适应(推理)两个时间尺度。他认为当前的 Transformer + 梯度下降 + token 空间思维链生成这一技术栈,效率甚至排不进前 50%。他断言未来的 AI 架构将与现在的技术栈截然不同。

Body

这篇观点延续了 Chollet 一贯的立场——他在 ARC-AGI 基准上长期质疑 LLM 的抽象推理能力上限,主张需要 beyond-token-space 的新机制。作为 Keras 创作者和前 Google 工程师,他的判断在架构研究方向上常被视为风向参考。

Original · François Chollet

I believe there is a vast, extremely diverse space of possible approaches to learning (for both time scales: long-term plasticity / pretraining & on-the-fly adaptation / reasoning), and that the current stack (Transformer + gradient descent & token-space reasoning chain generation) is not even in the top 50% most efficient.

Future AI will look nothing like the current stack.

Industry Coordinates & Source Verification

This report is categorized under AITOP's "Industry" desk, focusing on US frontier AI labs and Silicon Valley technology leaders. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include François Chollet, Transformer, ARC-AGI, reasoning models.

Source attribution: curated from François Chollet (@fchollet).