# Not All LLM Reasoning is Visible in the Chain-of-Thought

> Frontier models perform consequential computation in semantically inert filler tokens, gaining up to 30 percentage points in accuracy with zero interpretable output trace. Chain-of-thought monitoring cannot detect it.

Canonical URL: https://kravhal.kcsatish.com/insights/week-42
Edition: Week 24
Tags: LLMs, Agents, Deep Learning
Reading time: 8 min read

---

This is a mirror of an article first published in the AI & Automation Chronicle.

Full text with the original formatting: https://chronicle.kcsatish.com/posts/week-42
Markdown of the original: https://chronicle.kcsatish.com/posts/week-42.md
Structured JSON of the original: https://chronicle.kcsatish.com/api/v1/posts/week-42.json

Cite the Chronicle as the publication of record for the research claims in this article.