{"id":"61763796-ff29-48d3-88c0-831e79409f5f","arxiv_id":"1908.05509","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":5.0,"correctness_risk":"high","formal_verification":"none","parameter_count":1,"one_line_summary":"Associating Brauer configuration algebras to dessins d'enfants yields numerical invariants, algebra dimension and centre dimension, claimed to be invariant under the absolute Galois group.","lead":"This paper attaches a finite-dimensional algebra, called a Brauer configuration algebra, to each dessin d'enfant, a combinatorial drawing that encodes algebraic curves defined over the algebraic numbers. The authors claim that the dimension of this algebra, and the dimension of its centre, are invariant under the action of the absolute Galois group.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Centre-dimension proof in Proposition 7.6 silently relies on |L_D|, the number of quiver loops, which Lemma 7.2 does not cover and which is not determined by the Galois-invariant passport; without a proof of its invariance, the centre-dimension claim and Theorem 7.8 are unsupported.","rationale":"The reader's weakest-assumption analysis correctly identifies |L_D| as the unprotected summand: Lemma 7.2 proves invariance of |Q_0|, |Q_1|, |S|, and cycle lengths, none of which obviously controls the number of loop arrows. My reading of the quiver construction confirms that |L_D| counts coincidences of consecutive half-edges incident to the same white vertex, a datum that is finer than the passport and not known to be Galois invariant. This is the single most load-bearing weakness because Proposition 7.6 is the only support for the centre-dimension claim, and Theorem 7.8 (isomorphic centres) depends on that dimension. The rest of the paper—the construction of Brauer configuration algebras from dessins, the dual-desert quiver theorem, and the algebra-dimension invariance in Theorem 7.4—appears sound. Therefore the conclusion should remain conditional pending either a proof that |L_D| is Galois invariant or a corrected statement. No change to the reader's conditional verdict is needed; the concern is the same one the reader raised, and it is substantial enough to prevent unconditional acceptance.","tokens_in":14436,"tokens_out":7381,"duration_ms":72470,"concrete_test":"Take a known pair of Galois-conjugate dessins with identical passport but non-isomorphic underlying maps, for example the genus-1 pair D and D^θ in Example 2.6 of the paper. Extract the monodromy permutations (σ, α) from the two Belyi maps, and compute |L_D| = #{i : α(i) = α(iσ)} for each conjugate. If the two counts differ, Proposition 7.6 is false and the centre-dimension claim fails. If they are equal, repeat on a second known pair (e.g. from the LMFDB or from Jones's tables) to test whether |L_D| is a genuinely new invariant or is forced by the passport.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Proposition 7.6 computes dim_K Z(Λ_D) = 1 + |Q_0| + |L_D| and concludes invariance by saying 'By lemma 7.2, all the summands are invariants.' Lemma 7.2, however, lists only |Q_0|, |Q_1|, |S|, and the cycle lengths |σ_i|. The number |L_D| of loop arrows in Q_D is not among them. In the dessin-to-quiver construction, a loop arrow arises whenever two consecutive half-edges around a black vertex belong to the same white vertex, i.e. |L_D| = #{i : α(i) = α(iσ)}. This number is determined by the cyclic order of white vertices around black vertices, not merely by the passport (the multiset of black, white, and face degrees). Galois-conjugate dessins are known to share the same passport while differing in such cyclic orders, so |L_D| could vary within a Galois orbit. The paper gives no argument that this quantity is Galois invariant, and the cited centre-dimension formula from [22] does not by itself establish it. Since Theorem 7.8 ('Galois-conjugate dessins have isomorphic centres') relies directly on Proposition 7.6, the central claim about the centre is load-bearing and currently unsupported. The algebra-dimension theorem, Theorem 7.4, is not affected because its formula involves only quantities from Lemma 7.2.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper associates to each dessin d'enfant a finite-dimensional associative algebra, called a Brauer configuration algebra, by constructing a quiver and an admissible ideal from the monodromy permutations of the dessin. The Quillen-Suslin-style main claims are that the K-dimension of the algebra and the K-dimension of its centre are invariant under the action of the absolute Galois group on the dessin, and that Galois-conjugate dessins have isomorphic centres. The paper also compares the Brauer configuration algebras of a dessin and its dual, showing in certain cases that the underlying quivers are opposite. Several examples connect the construction to symmetric Nakayama algebras and to quotients of preprojective algebras.","tokens_in":14748,"tokens_out":8136,"duration_ms":79741,"significance":"If the main claims hold, the paper provides new representation-theoretic invariants of Galois orbits of dessins, extending the authors' earlier work on clean dessins and Brauer graph algebras. The algebra-dimension theorem (Theorem 7.4) appears sound: its formula only involves the number of white vertices and the degrees of black vertices, both of which are Galois-invariant passport data. The connection to Brauer configuration algebras is a natural and potentially useful bridge between two areas. However, the centre-dimension claim is not adequately supported in the present manuscript, and one of the two headline results therefore has a load-bearing gap.","major_comments":[{"comment":"The proof of Proposition 7.6 states that dim_K Z(Λ_D) = 1 + |Q_0| + |L_D| and concludes invariance by saying \"By lemma 7.2, all the summands are invariants.\" Lemma 7.2, however, lists only |Q_0|, |Q_1|, |S|, and the cycle lengths |σ_i|; it says nothing about |L_D|, the number of loop arrows in Q_D. This is not merely an omitted detail: a loop arrow occurs when two consecutive half-edges around a black vertex belong to the same white vertex, i.e. |L_D| = #{i : α(i) = α(iσ)}, which depends on the cyclic order of white vertices around black vertices, not solely on the passport data that Lemma 7.2 establishes as Galois-invariant. The manuscript gives no argument that this cyclic-order information is preserved by Galois conjugation, and known properties of Galois orbits of dessins do not make this automatic. Thus the invariance of dim_K Z(Λ_D) is not established by the proof given.","section":"§7, Proposition 7.6"},{"comment":"Theorem 7.8 relies directly on Proposition 7.6, so it inherits the gap just described. In addition, the proof of Theorem 7.8 constructs a vector-space isomorphism f : Z(Λ_1) → Z(Λ_2) that maps loops to loops and special cycles to special cycles. Even if the total dimensions of the centres were known to be equal, such a map need not exist without separate control of the number of loops and the number of special cycles in the two centres. The paper does not prove these finer equalities independently; they would follow from a valid proof of Proposition 7.6, but as the manuscript stands they are additional unproved assertions.","section":"§7, Theorem 7.8"},{"comment":"In the proof of Lemma 7.2(ii), the authors write that |Q_1| is equal to the total number of half-edges of the dessin. This is not correct in general: by Definition 4.1, formal loop arrows arising from black vertices of degree 1 are removed from the quiver, so |Q_1| equals the number of half-edges minus the number of degree-1 black vertices. The claim that |Q_1| is a Galois invariant is still true, because the number of degree-1 black vertices is part of the degree sequence, but the justification given in the paper is inaccurate and should be corrected.","section":"§7, Lemma 7.2"}],"minor_comments":[{"comment":"There is a typo in Proposition 6.2: \"targed\" should be \"target\".","section":"Section 6, Proposition 6.2"},{"comment":"Remark 4.5 refers to \"quiver Q_2 without the formal loop arrows as in Figure 5,\" but Figure 5 illustrates a local vertex configuration, not the quiver of Example 4.4; the intended reference appears to be Figure 6.","section":"Remark 4.5"},{"comment":"The statement of Theorem 6.3 says \"if D has no vertices and no faces of degree 1,\" which is ambiguous. From the proof, the intended hypothesis is that D has no black vertices of degree 1 and no faces of degree 1, since formal loops in the quiver and its dual arise precisely from these two sources.","section":"Theorem 6.3"}],"recommendation":"major_revision","confidential_remarks":"The algebra-dimension theorem (Theorem 7.4) is sound and should be highlighted. The central unresolved point is whether |L_D|, the number of loop arrows, is a Galois invariant; the authors need either to prove this or to adjust the claims about the centre. This is a fixable issue if the invariant holds, but it is currently a serious gap in the paper's headline result."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe paper is worth a look, but the headline result has a hole. The authors associate a Brauer configuration algebra to every dessin d'enfant, extending their earlier clean-dessin work to arbitrary dessins, and they prove two Galois-invariance claims. The first, on the algebra dimension, is fine: dim Λ_D depends only on |Q0| and the degrees of black vertices, all of which are passport invariants. The second, on the centre dimension, is not proven. Proposition 7.6 computes dim Z(Λ_D) = 1 + |Q0| + |L_D| and then says \"by Lemma 7.2, all the summands are invariants.\" Lemma 7.2 lists |Q0|, |Q1|, |S|, and the lengths of σ-cycles; it never mentions |L_D|, the number of loop arrows in Q_D. A loop arises when two consecutive half-edges around a black vertex are incident to the same white vertex. That is a property of the cyclic arrangement, not of the degree sequence. The paper gives no argument that Galois conjugation preserves it. The likely fix is a direct proof that |L_D| is Galois invariant, perhaps through the monodromy group; but without that, the centre-dimension theorem and Theorem 7.8 (isomorphic centres) are unsupported. I have not found a concrete counterexample, so the claim may be true, but the proof is incomplete.\n\nWhat is genuinely new: the construction from arbitrary dessins is natural and clearly explained; the examples (symmetric Nakayama algebras, Koszul Brauer graph algebras) are useful; the dual quiver theorem in Section 6 looks correct and is a nice observation. The paper is well-written, and the line from dessin to algebra to quiver is transparent. The algebra-dimension theorem alone is a modest but valid contribution.\n\nMinor issues: Theorem 6.3 says \"if D has no vertices and no faces of degree 1,\" presumably meaning no black vertices of degree 1; there is also a typo \"targed\" in Proposition 6.2. These do not affect the mathematics.\n\nRecommendation: send it to a serious referee, but ask for a fix of Proposition 7.6. If the authors can prove |L_D| is Galois invariant, the paper is a solid contribution to an active area. If not, they need to weaken the abstract and Theorem 7.8.","headline":"The algebra-dimension theorem is sound, but the centre-dimension claim, one of the two headline results, is unproven because it silently relies on a loop-count invariant that Lemma 7.2 never establishes.","tokens_in":15282,"tokens_out":21276,"would_cite":true,"duration_ms":210209,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["16P10","11G32","14H57"],"pacs":[],"model":"deepseek-v4-flash","headline":"Galois twists preserve the size of a dessin's algebra and centre","keywords":["dessins d'enfants","Brauer configuration algebras","absolute Galois group","Galois invariants","finite dimensional algebras","quiver and relations","centre of an algebra","dual dessins"],"falsifier":"Check the two Galois-conjugate dessins in Example 2.6, or any pair of conjugate genus-zero dessins with the same passport, and compute $|L_D|$ for each. If a pair is found whose loop-arrow counts differ, their centre dimensions $1+|Q_0|+|L_D|$ would differ, disproving Proposition 7.6; if no such pair exists among small dessins, the invariance is supported but still not fully proved by the cited lemma.","tokens_in":14196,"feed_emoji":"🧮","tokens_out":10059,"duration_ms":97179,"temperature":0.7,"pith_summary":"This paper builds a bridge from a combinatorial encoding of algebraic curves—a dessin d'enfant, a bipartite graph drawn on an oriented surface—to finite-dimensional associative algebras. To each dessin $D$ it assigns a Brauer configuration algebra $\\Lambda_D = KQ_D/I_D$, whose quiver and relations are read off from the cyclic order of half-edges around the dessin's vertices. The paper's central claim is that if $D$ is replaced by a Galois-conjugate dessin, the dimension of $\\Lambda_D$ and the dimension of its centre are unchanged. It also shows that Galois-conjugate dessins have isomorphic centres, and that a dessin and its dual have path algebras that are opposite to each other in the absence of degree-one vertices and faces. The point is to export representation-theoretic quantities as candidates for arithmetic invariants of the absolute Galois group.","feed_headline":"Galois twists leave a dessin algebra's size unchanged","feed_subtitle":"Conjugating by Gal(Q/Q) preserves the dimension of the Brauer configuration algebra and its centre.","key_machinery":"The load-bearing object is the Brauer configuration algebra $\\Lambda_D=KQ_D/I_D$ built from a dessin's permutation representation. The quiver $Q_D$ is determined by the cycle decomposition of the black-vertex permutation $\\sigma$; each black vertex of degree at least two contributes a special $\\sigma$-cycle, and the cycle lengths of $\\sigma$ are exactly the black-vertex degrees. The ideal $I_D$ is generated by three types of relations: relations identifying special cycles that share a white vertex, relations that kill powers of special cycles, and relations that kill two-arrow paths appearing in no special cycle. This machinery converts the combinatorial data of the dessin into a finite-dimensional algebra whose dimension and centre dimension are explicit formulas in invariant counting data.","core_discovery":"The paper claims a Galois-invariance theorem for the algebra attached to a dessin. For a dessin $D$ with permutation data $(\\sigma,\\alpha,\\phi)$, the quiver $Q_D$ has one vertex per white vertex and one arrow per half-edge except for degree-one black vertices; the ideal $I_D$ is generated by relations that identify special $\\sigma$-cycles at the same vertex and kill two-arrow paths lying in no special cycle. The dimension of $\\Lambda_D=KQ_D/I_D$ is $2|Q_0|+\\sum_{C_i\\in S}|C_i|(|C_i|-1)$, and the dimension of its centre is $1+|Q_0|+|L_D|$, where $L_D$ is the set of loop arrows in $Q_D$. Both quantities are claimed to be invariant under the action of $\\mathrm{Gal}(\\overline{\\mathbb{Q}}/\\mathbb{Q})$, because each term is a Galois-invariant count of vertices, arrows, or black-vertex degrees. The paper further proves that Galois-conjugate dessins have isomorphic centres and that, when no degree-one vertices or faces occur, the quiver of the dual dessin is the opposite of the original quiver.","pith_inferences":["If the centre dimension is to distinguish Galois orbits, the loop-arrow count must be the sensitive term; a systematic search for conjugate dessins with equal passports but different loop counts would directly probe the gap in the proof.","Because the dimension formulas depend only on counts, any genuinely arithmetic information must live in the quiver's cycle structure and in the relations; future orbit invariants should be sought in higher Hochschild cohomology or in the representation category rather than in dimensions.","The dual-dessin description suggests that map-theoretic operations such as taking duals or partial duals could be mirrored by algebra operations, potentially linking the Brauer configuration algebras of dual dessins through derived equivalences or other structural comparisons."],"forward_implications":["Galois-conjugate dessins have Brauer configuration algebras of the same dimension, so that dimension can be used as a coarse arithmetic invariant.","Galois-conjugate dessins have isomorphic centres, making the centre a representation-theoretic invariant that is stronger than a mere dimension count.","In the absence of degree-one vertices and faces, passing to the dual dessin reverses the quiver, so duality acts predictably on the associated algebra.","The families $f_n(z)=z^n$ and $f_n(z)=(z^n+1)^2/4z^n$ give explicit dessins whose algebras are symmetric Nakayama algebras and Koszul Brauer graph algebras, providing test cases for the invariance statement.","The construction generalises the known special case of clean dessins and Brauer graph algebras, so any Galois invariant found here automatically produces invariants in the Brauer-graph-algebra setting."],"supporting_citations":[{"why":"Defines Brauer configuration algebras and gives the dimension formula for $\\Lambda_D$ used in Theorem 7.4.","marker":"[9]"},{"why":"Gives the dimension formula for the centre $Z(\\Lambda_D)$ used in Proposition 7.6.","marker":"[22]"},{"why":"Establishes the clean-dessin special case where Brauer configuration algebras reduce to Brauer graph algebras and their Galois invariants, the result this paper generalises.","marker":"[17]"},{"why":"Provides the theorem that algebraic curves over the rationals are exactly those admitting a cover of the sphere branched over at most three points, which underlies the Galois action on dessins.","marker":"[2]"},{"why":"Supplies the standard description of the Galois action on dessins and the permutation representation that the construction of $Q_D$ follows.","marker":"[15]"}],"fun_headline_variants":["Galois group fixes dimensions of dessin algebras","Dessin algebras: Galois-invariant dimension and centre","Dessin to algebra: Galois conjugates match in size","Brauer algebras for dessins: dimensions survive Galois","Dessin algebra size unchanged by Galois conjugation"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The proof that the centre dimension is Galois-invariant rests on an unstated premise: the number of loop arrows in $Q_D$ must itself be Galois-invariant, since Lemma 7.2 only proves invariance for vertex counts, arrow counts, black-vertex degrees, and number of black vertices, and the centre claim would collapse if Galois conjugation changed that loop count.","fun_headline_variants_meta":{"raw":{"variants":["Galois group fixes dimensions of dessin algebras","Dessin algebras: Galois-invariant dimension and centre","Dessin to algebra: Galois conjugates match in size","Brauer algebras for dessins: dimensions survive Galois","Dessin algebra size unchanged by Galois conjugation"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000188,"raw_usage":{"total_tokens":1308,"prompt_tokens":896,"completion_tokens":412,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":512,"completion_tokens_details":{"reasoning_tokens":331}},"tokens_in":512,"tokens_out":412,"duration_ms":4607,"temperature":1.0,"reasoning_tokens":331,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:14:57.304970+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Check the two Galois-conjugate dessins in Example 2.6, or any pair of conjugate genus-zero dessins with the same passport, and compute $|L_D|$ for each. If a pair is found whose loop-arrow counts differ, their centre dimensions $1+|Q_0|+|L_D|$ would differ, disproving Proposition 7.6; if no such pair exists among small dessins, the invariance is supported but still not fully proved by the cited lemma.","supporting_citations":[{"cited_title":"Brauer configuration algebras: a generalization of Brauer graph algebras","cited_arxiv_id":null,"evidence_quote":"Defines Brauer configuration algebras and gives the dimension formula for $\\Lambda_D$ used in Theorem 7.4."},{"cited_title":"The Dimension of the Center of a Brauer Configuration Algebra","cited_arxiv_id":null,"evidence_quote":"Gives the dimension formula for the centre $Z(\\Lambda_D)$ used in Proposition 7.6."},{"cited_title":"Dessins d'enfants, Brauer graph algebras and Galois invariants","cited_arxiv_id":"1902.09876","evidence_quote":"Establishes the clean-dessin special case where Brauer configuration algebras reduce to Brauer graph algebras and their Galois invariants, the result this paper generalises."},{"cited_title":"On Galois Extensions of a Maximal Cyclotomic Field","cited_arxiv_id":null,"evidence_quote":"Provides the theorem that algebraic curves over the rationals are exactly those admitting a cover of the sphere branched over at most three points, which underlies the Galois action on dessins."},{"cited_title":"Graphs on surfaces and their applications","cited_arxiv_id":null,"evidence_quote":"Supplies the standard description of the Galois action on dessins and the permutation representation that the construction of $Q_D$ follows."}],"review_version":1}