# Transformer学习局限与RL的突破潜力

- 来源：swyx (@swyx)
- 发布时间：2026-05-23 14:33
- AIHOT 分数：58
- AIHOT 链接：https://aihot.virxact.com/items/cmphzh45g0ssasljwurom9ifs
- 原文链接：https://x.com/swyx/status/2058073815301972368

## AI 摘要

本文肯定了对Transformer当前学习能力及局限性的分析框架，并指出对抗性世界模型是逼近现实本质的关键功能之一。作者认为，单纯增加参数和算力以扩展一个低效范式，将被能主动假设与验证真理的简洁方案所超越，尽管规模化可能因人类智能本身有限而意外通向AGI。引用推文补充了强化学习（RL）作为从干预中学习的范式，比监督学习更强大，而世界建模与RL的结合有望实现对反事实的学习。

## 正文

co-sign. a very handy mental framework for what kinds of learning transformers do well today， and why it runs into limitations. when @ankit2119 and i wrote about the need for adversarial world models earlier this year， we were describing a couple of the functions of these rungs of thinking that bring us ever closer to the kolmogorov-limit generator of reality. throwing more params， more power， more everything at a demonstrably inefficient paradigm will be outclassed by the simple solution that can hypothesize and seek truth rather than backfit a house of cards - although the bitter lesson is it is simpler to scale and we may hit agi anyway because human intelligence just isn't that smart nor plentiful

### 引用推文

> Rishabh Agarwal：Very well written blog. I think of RL as learning from interventions, and it kinda explains why it's more powerful as a paradigm than supervised learning. Now l...