Third Brauer–Thrall conjecture

A module is indecomposable if it cannot be split into a direct sum of two nonzero modules. If a finite-dimensional algebra has infinitely many types of finite-dimensional indecomposable modules, must it also have indecomposable modules of arbitrarily large infinite dimension? This is the third Brauer–Thrall conjecture, also called Simson’s conjecture.

Why it matters and what counts as progress

Why it matters: An indecomposable module is a building block for representations of an algebra. The conjecture asks whether having infinitely many finite-dimensional building blocks forces building blocks at every sufficiently large scale.

What counts as progress: Construct indecomposable modules of unbounded dimension for a new class of algebras or fields. A stronger result can prescribe the exact dimension, count isomorphism classes or control the rings of module endomorphisms.

How the problem developed

  1. 2023 · Finite-dimensional embeddings [1]

    Klaus Bongartz Klaus Bongartz constructed representation embeddings from a category of finite-dimensional Kronecker modules. These embeddings transfer indecomposable modules to a representation-infinite algebra over an algebraically closed field.

  2. 4 Nov 2026 · Embeddings in arbitrary dimension [2]

    Haruhisa Enomoto with GPT-6 Astra Haruhisa Enomoto extended the embedding to modules of arbitrary dimension. He credits GPT-6-Astra with developing the proof. The result gives indecomposable modules of each infinite dimension over algebraically closed fields.

AI contribution

Representation embeddings and the third Brauer–Thrall conjecture

Haruhisa Enomoto

Let A be a representation-infinite finite-dimensional algebra over an algebraically closed field k. Enomoto constructs a representation embedding from Kronecker modules without a simple injective summand into A-modules. The embedding preserves each infinite dimension. For every infinite cardinal κ\kappa , the paper obtains 2κ2^{\kappa } nonisomorphic indecomposable A-modules of dimension κ\kappa with residue field k and endomorphism radical nilpotence exponent at most max⁡(30,4dim⁡A)\max (30, 4 \dim A).

AI involvementAI-led
ImportanceNotable
Technical verificationFormal (Lean)
Human understandingFew
Paper date:

Sources

  1. Bongartz’s finite-dimensional embedding theorem
  2. Representation embeddings and the third Brauer–Thrall conjecture
  3. Lean proof and paper correspondence
Publication history