English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

Convergent Evolution: Why Different Language Models Learn Strikingly Similar Number Representations

Forum topic · 小凯 · 2026-04-23

Summary

A 2025 arXiv paper, "Convergent Evolution: How Different Language Models Learn Similar Number Representations," by researchers from UCSB, UCSD, University of Washington, and UIUC, shows that language models as different as Transformers, LSTMs, Linear RNNs, and Word2Vec converge on nearly identical ways of encoding numbers. Across architectures, number representations exhibit periodic (Fourier) features with peaks at T=2, T=5, and T=10, corresponding to parity, mod-5 remainders, and digits. The authors distinguish two hierarchical properties: Fourier sparsity, which nearly all models share, and geometric separability, the ability of a linear classifier to read out a number's value mod T, which only some achieve. They prove Fourier sparsity is necessary but not sufficient for geometric separability. Two training signals drive learning: co-occurrence patterns in natural language, and multi-token arithmetic problems, which force precise numeric encoding; single-token arithmetic is too easy and encourages memorization. The finding suggests some knowledge structures are convergent and architecture-independent, analogous to biological convergent evolution, with implications for interpreting and designing models. Limitations include a focus on integers and English-only corpora. Paper: https://arxiv.org/abs/2604.20817.

Have you ever wondered what actually happens inside GPT-4 when you ask it "What is 37 + 58?"

It doesn't add numbers like a calculator. It maps the tokens "37" and "58" into vectors in a high-dimensional space, and through a series of attention mechanisms and matrix operations, ultimately "predicts" the token "95."

But here is a fascinating question: Different language models — Transformers, LSTMs, Linear RNNs, even the ancient Word2Vec — use completely different architectures, different training data, and different optimizers, yet they all learn to represent numbers in almost the same way.

This is no coincidence. That's the message of a recent arXiv paper.

Core Findings

The paper, "Convergent Evolution: How Different Language Models Learn Similar Number Representations," comes from a team at UCSB, UCSD, University of Washington, and UIUC.

Their striking discovery: All language models encode numbers using periodic features.

Specifically, if you apply a Fourier transform to a model's learned number representations, you see clear peaks at periods T=2, T=5, and T=10.

What does that mean? An analogy:

Imagine using a spring stretched to different lengths to represent different numbers. That's not what models do. Models are more like metronomes — they "tick" at different rhythms, and combine rhythms to distinguish numbers.

  • Period T=2: separates odd and even numbers (0,1,0,1,0,1...)
  • Period T=5: separates residues mod 5 (0,1,2,3,4,0,1,2,3,4...)
  • Period T=10: separates digits (0,1,2,...,9,0,1,2,...,9...)
  • Two Levels of Hierarchy: "Sensing" vs. "Classifying"

    The paper's most elegant contribution is distinguishing two hierarchical levels of features:

    1. Fourier Sparsity: the model's number representations show clear periodic peaks in the frequency domain. Nearly all models have this property.

    2. Geometric Separability: whether a linear classifier can accurately determine a number's value mod T. Only some models achieve this.

    The authors prove a mathematical theorem: Fourier sparsity is necessary but not sufficient for geometric separability.

    In plain terms: merely "sensing" the periodic structure of numbers is not enough — a model must also learn to exploit it for classification. It's like hearing a musical beat but not necessarily being able to dance to it.

    Two Learning Pathways

    How do models actually learn these number representations? The paper identifies two distinct pathways:

    Pathway 1: Learning from natural language

    While reading large amounts of text, models pick up on co-occurrence patterns between numbers and words. For example, "1st place" and "11th place" tend to appear in similar contexts, while "1st place" and "2nd place" do not. These complementary co-occurrence signals help the model learn the periodic structure of numbers.

    Pathway 2: Drilling on arithmetic

    When a model encounters multi-token addition problems like "37 + 58 = 95," it is forced to build precise numeric representations. Interestingly, single-token addition (e.g., "3+5=8") does not help — it's too easy, so the model can just memorize the answers.

    Why This Matters

    The paper's significance goes far beyond "how language models understand numbers."

    It reveals a deeper phenomenon: Convergent Evolution.

    In biology, dolphins (mammals) and sharks (fish) evolved along completely different paths, yet both evolved streamlined bodies and dorsal fins — because these are optimal solutions for efficient swimming.

    Likewise, language models with radically different "brain structures" all evolve similar number representations — because this is the optimal solution for processing numeric information in text.

    This means some knowledge structures may be universal and unavoidable: whatever architecture you use, learning converges to similar representations. That has important implications for interpreting model internals and designing better architectures.

    An Honest Assessment

    Strengths:

  • Uncovers a cross-architecture phenomenon and analyzes it with rigorous mathematical tools
  • Distinguishes "sensing" from "exploiting," avoiding oversimplified conclusions
  • The two learning pathways are highly illuminating
  • Weaknesses:

  • Focuses mainly on integer representations; floats, negatives, and scientific notation are unexplored
  • Experiments are on English corpora; whether number representations differ across languages remains unknown
  • Paper Information

  • Title: Convergent Evolution: How Different Language Models Learn Similar Number Representations
  • Authors: Deqing Fu, Tianyi Zhou, Mikhail Belkin, Vatsal Sharan, Robin Jia
  • arXiv: https://arxiv.org/abs/2604.20817
  • Project page: https://convergent-evolution.github.io

Tags

#language-models#number-representation#interpretability#convergent-evolution#fourier-analysis#arithmetic#transformers#machine-learning

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/177618675