Monna Map Generation and Hallucination

In the paper's words

Every thought you have—a memory, a plan, a sudden insight—arrives not as a neat list of words but as a living structure. It is a branching hierarchy of concepts, where broad categories contain finer details, and relationships are defined by containment and connection rather than sequence. This mental architecture is not merely a metaphor; it has a precise mathematical counterpart known as an ultrametric tree.

The abstract, verbatim, and its opening sentence is the claim. Every line below is quoted from the text.

Numbers it reports

No sentence of this paper reports a figure with a unit or a quantity word, so none is shown.

How it is done

No section of this paper is shaped like a method, so no method sentence is quoted.

What would falsify it

1. The same hierarchical clustering emerges in the latent spaces of large language models: gradient descent, optimizing to predict the next token across trillions of examples, forces the model’s internal representations to organize ultrametrically.

   from 1.1 The

2. This parallel reveals a deep structural unity: whenever a rich, multi‑dimensional state is collapsed onto a low‑dimensional observable, information is sacrificed.

   from 4. The Quantum

Where the claim stops

1. When your brain parses a sentence, it does not treat the words as a flat string.

   from 1.1 The

2. Because trailing zeros in the p -adic expansion (which correspond to “no further detail” down a branch) become leading zeros in the real expansion, and leading zeros do not change the value of a real number.

   from 3.1 The Many‑to‑One Nature of

3. However, multiple distinct tree‑paths can project to the same token sequence.

   from 3.2 The Origin of

Where to read

1. A264 words · 1 min
2. 1.517 words · 2 min
3. 2. The Monna Map: From Tree to572 words · 2 min
4. 3. Hallucinations514 words · 2 min
5. 4. The Quantum173 words · 1 min
6. 5. Implications for335 words · 1 min
7. The Shape of Meaning177 words · 1 min