# CMU与牛津：循环流让小模型靠隐状态迭代推理

- 来源：Rohan Paul (@rohanpaul_ai)
- 发布时间：2026-09-18 08:43
- AIHOT 分数：33
- AIHOT 链接：https://aihot.news/items/cmu69kfa20g2profjzclf8zb5
- 原文链接：https://x.com/rohanpaul_ai/status/2100747487682449894

## AI 摘要

卡内基梅隆大学与牛津大学论文提出"循环流"（looped flows）方法，让模型通过反复改进隐状态而非生成更长的思维链来"思考更久"，额外迭代步数可显著提升准确率，小模型也能借此获得更强推理能力，意味着测试时计算不必等同于生成更多 token。该方法通过让每次更新学习一个小型去噪任务、同时保证隐状态对下一次更新仍有用，解决了循环模型长循环训练不稳定的问题。

## 正文

New Carnegie Mellon + Oxford Univ paper shows a model can "think longer" by repeatedly improving hidden state rather than writing a longer chain-of-thought.

and looped flows show those extra steps can materially raise accuracy.

that small models can reason better by refining hidden state for more steps, so test-time compute does not have to mean generating more tokens.

The problem is that recurrent models are hard to train over long loops: the model may keep updating its hidden state, but those updates can become unstable or stop helping.

Looped flows fix this by training each update on a small denoising task while making sure the hidden state remains useful for the next update.

This lets the model keep improving the same internal representation at inference time.