The emergence of Jev has once again dug up the so-called
A commenter framed Jev as carving typed cognitive operations out of LLM language generation, reviving the System 1 versus System 2 debate.
The original post, translated
The emergence of Jev has once again dug up the so-called sys1 and sys2 capabilities that people debated when LLMs first appeared. The LLM based on NTP itself is a symbolic generation process biased toward intuitive sys1 representation, and later we developed reasoning models like o1 based on CoT, adding its sys2 process. Jev cuts out typed cognitive operations from the language generation process of LLMs. Taking what was originally forced through
Show the original post in its source language ▾
Jev的出现，又翻出了LLM出现之初，人们争议的所谓sys1和sys2能力。 基于NTP的LLM本身是一个偏直觉sys1表示的符号生成过程，后来我们又基于CoT发展出o1这样的推理模型，增加了其sys2的过程。 Jev 从 LLM 的语言生成过程中，切出了可类型化的认知操作（typed cognitive operations）。 把原本被迫通过
The post above is a machine translation from zh; the untranslated text is in the fold-out.
Engagement when collected
Numbers are a snapshot taken from X when the case was added to the library (schema v1, collected 2026-09-19); they will not match today.
Where this case fits
Filed under general Jev builds. In the pattern Jev is built for, the model answers a bounded question per step — and ordinary code acts on the answer, because the answer is already a value rather than a paragraph. Browse the full case library for the work it sits next to.
Related Jev cases
MoE is: doing specialization in parameter space.
There is still some difference between the sys1 here and a
Keep browsing: all 1173 Jev cases · more from @TaNGSoFT · builders · what Jev is
Last updated: 2026-09-22 · sources & corrections · every card links to its author's original post