Results
on bricks in representation theory of algebras
This page collects new results on bricks — partial developments, ideas and promising directions of work that have been communicated to the Brick Lodge — with full credit to those who found them. The entries are listed in the order in which they were added, and may not yet have appeared in a refereed journal or on the arXiv.
Some of the results below may be AI-assisted, or even entirely AI-generated, under a mathematician’s direction. They are shared solely to help build intuition and insight into the open problems of the Brick Lodge and related topics. If you wish to cite or use any of them in a publication or presentation, please contact the contributors directly. Each result is credited to the person or people who communicated it.
The Brick Lodge welcomes feedback and comments from all experts on these results, as well as suggestions for improving their exposition (with credit to the original work). To leave a comment, use the “Comments & Feedback” button at the bottom right of the relevant result. Once the maintainer or one of the contributors of the Brick Lodge has reviewed and approved it, your comment will appear beneath that result under your full name.
1. On E-Symmetric Algebras and brick-Brauer–Thrall Conjectures
Setting: k is an algebraically closed field, and A = kQ/I is a basic, connected, finite-dimensional k-algebra; modules are finite-dimensional left modules. Following Bobiński–Schröer (arXiv:2610.17070), A is E-symmetric if dimk HomA(M, τM) = dimk HomA(τ−1M, M) for every M in mod A.
Further notation (used in Theorems B and C): A is minimal brick-infinite if it has infinitely many bricks while every proper quotient of A is brick-finite. For a weight θ in K0(proj A), a nonzero module B is θ-stable if θ(dim B) = 0 and θ(dim L) < 0 for every nonzero proper submodule L of B. Let ℱA be the support of the τ-tilting fan (the union of the g-vector cones of the support τ-tilting modules), in Rn with n the rank of A. Then A is g-tame if the closure of ℱA is Rn, and E-finite if Qn ⊂ ℱA (otherwise E-infinite). Finally CA = (dimk eiAej) is the Cartan matrix, qA(x) = xTCAx, and gT denotes the g-vector of a τ-rigid module T, with ĝT = gT/‖gT‖1.
Theorem A (Alternative characterization). For M in mod A, let gM = [P0] − [P1] and hM = [I0] − [I1] be computed from the minimal projective presentation P1 → P0 → M → 0 and the minimal injective copresentation 0 → M → I0 → I1, viewed as vectors in ZQ0, and let BQ = (qij − qji), where qij is the number of arrows i → j in Q. Then A is E-symmetric if and only if gM − hM = BQ dim M for every M in mod A.
Theorem B (Stable Second brick-Brauer–Thrall). Let A be E-symmetric and minimal brick-infinite. Then A is E-infinite, and there exist a nonzero dimension vector b in Z≥0n and an integral weight θ in K0(proj A) for which A has infinitely many pairwise non-isomorphic θ-stable bricks of dimension vector b.
Theorem C (τ-convergence property). Let A be g-tame, E-symmetric and minimal brick-infinite. Then there exist pairwise non-isomorphic indecomposable faithful τ-rigid modules Tj with pdA Tj = 1 such that qA(ĝTj) = dimk EndA(Tj) / ‖gTj‖12 → 0 as j → ∞. In particular, A has the τ-convergence property, with an actual convergent sequence of indecomposable modules.
Credit, background and motivation. These results were obtained by Kaveh Mousavand in a supervised AI-investigation with OpenAI’s GPT-6 (the attached note lists “Chat GPT-6 Sol” as a co-author), on three occasions: October 14, November 11 and November 12, 2026. The note, dated November 16, 2026, was edited and polished by K. Mousavand, who makes no claim of novelty, priority or ownership over the statements it records; results quoted from the literature are distinguished in it from arguments supplied by the investigation. The note is not on the arXiv and has not been refereed, and the author welcomes corrections and suggestions. Theorems B and C come from the third session, and are the part of most interest for the Brick Lodge. The author thanks Charles Paquette for many years of collaboration on the brick-Brauer–Thrall conjectures.
Summary of the note. (1) E-symmetric algebras. Theorem A characterizes E-symmetry by the identity gM − hM = BQ dim M, and it follows that CATBQCA = CAT − CA. The note also proves that for a weakly triangular quiver (no oriented cycles apart from loops) an algebra is E-symmetric exactly when it is a directed free gluing of local E-symmetric algebras; in particular, for an acyclic quiver, A is E-symmetric if and only if A = kQ. It further shows that an E-symmetric algebra of rank n that is not strictly wild has at most 9n rigid bricks of finite projective dimension in any semibrick. (2) Brick-Brauer–Thrall conjectures. Theorems B and C: the Stable Second bBT Conjecture and the τ-convergence property hold for E-symmetric minimal brick-infinite algebras (for C, also g-tame). (3) Examples and non-examples. Affine hereditary algebras, generalized barbell algebras and a square-zero-loop algebra satisfy the hypotheses; the algebra Ω3 is g-tame and minimal brick-infinite but not E-symmetric, and has no τ-convergence. A 14-dimensional E-symmetric algebra is also constructed that is brick-infinite but has no E-symmetric minimal brick-infinite quotient.
E-symmetric algebras. For a module M, the number dim HomA(M, τM) is a basic invariant of M in τ-tilting theory (it vanishes exactly when M is τ-rigid), and the algebra is called E-symmetric when this invariant is symmetric with respect to the translation, in the sense of the displayed equality in the Setting. The notion was introduced by Bobiński–Schröer (arXiv:2610.17070) and developed in Bobiński–Hartung–Schröer (arXiv:2611.13889), where it is characterized by a condition on simple modules, by the 1-Iwanaga–Gorenstein property together with a duality, and by being a “generalized Jacobian algebra”. A local algebra is E-symmetric exactly when it is self-injective and its non-projective simple module, if any, has Ω4S ≅ S. Hereditary algebras of affine type are E-symmetric, and so are the algebras obtained by taking powers of loops as relations, via a hyperpotential. The class therefore contains the path algebras of the affine Dynkin quivers, and it is related to Jacobian algebras: the note recalls that the weakly symmetric algebras Aq on the Markov quiver are E-symmetric for q ≠ 0 but are not classical Jacobian algebras for q ≠ 1, in an example that the second paper draws from Enomoto’s counterexample to the periodicity conjecture (arXiv:2609.09732). Theorem A, as the note itself remarks, is the same as Proposition 3.13 of arXiv:2611.13889, after changing the signs of the presentation vectors used there. Please note that “E-symmetric” is a different notion from the properties “E-finite”, “E-infinite” and “E-tame” that occur elsewhere in the Brick Lodge, and should not be confused with them.
The brick-Brauer–Thrall conjectures. The brick-Brauer–Thrall (bBT) conjectures are the brick analogues of the classical Brauer–Thrall conjectures. They predict that brick-infinite algebras (equivalently, τ-tilting infinite algebras) have infinitely many bricks of a fixed dimension, as in the 2nd bBT Conjecture, or even infinitely many θ-stable bricks of a fixed dimension vector for a fixed weight θ, as in the Stability Conjecture. The latter is the statement called the Stable Second bBT Conjecture, which implies the former, because θ-stable modules are bricks. Both are closely tied to the g-vectors Conjecture, which says that a brick-infinite algebra has a rational ray outside the τ-tilting fan, that is, is E-infinite. For a minimal brick-infinite algebra, one can try to produce such a ray with the τ-convergence property of Mousavand–Paquette (arXiv:2311.14863): an infinite family of indecomposable τ-rigid modules whose g-vectors, once normalized, approach the null cone of the quadratic form qA. Such algebras are E-infinite, and an E-infinite algebra has a common-weight stable family of bricks by arXiv:2609.31417. Theorems B and C verify precisely this chain for E-symmetric algebras. Roughly speaking, the proof of Theorem C has two cases. If some τ-orbit of indecomposable τ-rigid modules is infinite, E-symmetry makes the dimensions of the endomorphism algebras constant along the orbit, while the g-vectors go to infinity, which forces qA of the normalized g-vectors to tend to 0. If all orbits are finite, the sum U of the modules in a periodic orbit gives a nonzero integral vector in the kernel of CA + CAT (so qA vanishes on it), which is shown to lie outside the τ-tilting fan using generic presentations (Plamondon, doi:10.1093/imrn/rns102); the g-tameness is then used to approach this vector by classical tilting cones and to extract a convergent sequence of indecomposables. The reduction to minimal brick-infinite algebras comes from the framework of the survey arXiv:2508.11789, and the case of algebras that are not g-tame is covered by the previous work on E-infinite algebras. The results do not settle the bBT conjectures for all E-symmetric algebras: the note shows that E-symmetry is not preserved by quotients, via a 14-dimensional algebra that is E-symmetric and brick-infinite but whose unique minimal brick-infinite quotient is not E-symmetric. The author also reports that the arguments should extend to weakly minimal brick-infinite algebras (minimality taken only with respect to quotients by ideals generated by idempotents), which would cover the algebras of Bongartz and Happel–Vossieck; this extension is still being verified and is not part of the note.
Connections to other entries of the Brick Lodge:
• 2nd bBT Conjecture, Stability Conjecture and g-vectors Conjecture: Theorem B verifies the first two for E-symmetric minimal brick-infinite algebras, and Theorem C supplies the τ-convergence property, hence an E-infinite algebra, as the third predicts. This adds to the families recorded there, but, as explained above, it is not known for all E-symmetric algebras.
• Generic-brick Conjecture, Semibrick Conjecture and Hom-orthogonal Conjecture: for the weakly symmetric algebras Aq (q ≠ 0) on the Markov quiver, which are E-symmetric, the note recalls that the two-dimensional modules Uλ (λ ≠ 0) are pairwise Hom-orthogonal bricks, stable for a common weight and fixed by τ. Hence Aq has an infinite semibrick of bricks of the same dimension, and a generic brick of endolength 2 (replace λ by an indeterminate); all this is consistent with these three conjectures. These bricks are not rigid: dim Ext1(Uλ, Uλ) = 1.
• Rigid-bricks Conjecture, Open Question 3 (Rigid semibricks) and Observation 1: the note proves that, for an E-symmetric algebra of rank n that is not strictly wild, every semibrick has at most 9n rigid bricks of finite projective dimension, and that in an infinite rigid semibrick all but at most 9n members have infinite projective and infinite injective dimension. The algebra of Observation 1 is strictly wild, so there is no tension with it. The bound is related to part (2) of Question 3, namely whether an algebra with an infinite semibrick of rigid bricks must be wild, but does not answer it.
• Open Question 1 (tameness): g-tameness is one of the four notions of tameness there, and it is used as a hypothesis in Theorem C. The results do not decide whether the notions coincide. We do not know of a direct implication for, or further insights into, the remaining questions and conjectures of the Brick Lodge, in particular Observation 2 and the length-category question of Conjecture 8.
Related topics
The comment form has not been activated yet. Please check back soon, or contact Kaveh Mousavand directly.