FluxEM — Algebraic Embeddings

The premise

We've been teaching neural networks arithmetic the hard way

For a decade we've scaled parameters and tuned attention mechanisms. Models can discuss calculus and explain quantum mechanics, yet they still stumble on multiplication.

It's not a capacity problem; it's representation. Tokenize "1234" into arbitrary symbols and you're asking the model to rediscover—through gradient descent—that those symbols have algebraic structure, again and again.

FluxEM takes the opposite tack: encode numbers so that vector addition is arithmetic addition. The structure is given up front, not learned.

And it doesn't stop at arithmetic: physical units, molecular formulas, musical pitch classes, logical propositions—any domain with algebraic structure can be embedded so its operations become geometric.

Eleven domains

Existence proofs for a general principle

Physics

9.8 m/s²

Dimensional quantities with unit conversion and dimensional analysis

Chemistry

C₆H₁₂O₆

Molecular formulas with stoichiometric operations and mass balance

Biology

ATGCCGTAG

DNA sequences with GC content, melting temperature, translation

Mathematics

3 + 4i

Complex numbers, matrices, vectors, polynomials, rationals

Logic

p ∧ q → r

Propositional and predicate logic with satisfiability detection

Music

{0, 4, 7}

Pitch classes, chords, scales, atonal set theory with TnI operations

Geometry

△ABC

Shapes with area, centroid, circumcenter, containment tests

Graphs

G = (V, E)

Adjacency structures with connectivity, cycles, shortest paths

Sets

A ∪ B

Finite sets and relations with union, intersection, composition

Number Theory

360 = 2³·3²·5

Prime factorization, modular arithmetic, Euler totient

Data

[x₁, x₂, ...]

Arrays, records, and tables as structured embeddings

Try it

Five minutes to working code

# Install
pip install fluxem            # or: pip install fluxem[jax]

# Use
from fluxem import create_unified_model

model = create_unified_model()
model.compute("12345 + 67890")   # → 80235.0
model.compute("144 * 89")        # → 12816.0

# Or work with specific domains
from fluxem.domains.music import AtonalSetEncoder, prime_form
from fluxem.domains.chemistry import MoleculeEncoder, Formula

enc = AtonalSetEncoder()
major = enc.encode([0, 4, 7])     # C major triad as pitch-class set
prime_form([0, 4, 7])            # → (0, 3, 7)

mol = MoleculeEncoder()
glucose = mol.encode(Formula.parse('C6H12O6'))