{"id":"e9ed61ad-e661-41fd-9d8b-2e51ed43468b","arxiv_id":"2509.16997","paper_version":2,"verdict":"REJECT","confidence":"HIGH","novelty_score":5.0,"correctness_risk":"high","formal_verification":"none","parameter_count":0,"one_line_summary":"The claimed j(g) formula and the associated complex multiplication coupling values are incorrect: Eq. (80) does not follow from Eqs. (78) and (79).","lead":"This paper derives a j-invariant for the spectral curve of a two-cut quartic Hermitian matrix model and claims five couplings where the curve has complex multiplication. The central formula is algebraically inconsistent with the paper's own cross-ratio equations, so the reported couplings are not supported.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Eq. (80) is not the substitution of Eq. (79) into Eq. (78); the correct j(g) is 256(1-3g)^3/[g^2(1-4g)], so every g in Table 1 is obtained from an algebraically false formula.","rationale":"The reader's weakest_assumption identifies exactly the same load-bearing algebra step. I independently re-derived it: the cross-ratio from the spectral edges is correct, but the claimed simplification to Eq. (80) is false. This is a decisive internal inconsistency for the paper's main table. I do not object to the background material or to the possibility of CM in spectral curves; the paper simply does not provide correct evidence for its specific claim. The verdict of REJECT is already based on this issue, so I leave it unchanged.","tokens_in":10012,"tokens_out":11896,"duration_ms":86795,"concrete_test":"Run a symbolic simplification (e.g., SymPy) of j(r(g)) from Eqs. (78)-(79); if it returns 256(1-3g)^3/[g^2(1-4g)] rather than Eq. (80), Table 1 is refuted. Independent numeric cross-check: at g=1/8, compute a^2=8+2/sqrt(g)=13.657, b^2=8-2/sqrt(g)=2.343, r=-(a-b)^2/(4ab)=-0.2071, and j from Eq. (78) equals 8000, whereas Eq. (80) gives 31.25.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Good-faith reading: the large-N two-cut solution up to Eq. (72) is standard, and the spectral curve y^2=(x^2-a^2)(x^2-b^2) is the right object. The load-bearing step is the Legendre conversion. With q=sqrt(1-4g), Eq. (79) gives r=(q-1)/(2q). Substitution into the standard j-invariant (78) yields: r(r-1)=g/q^2, r^2-r+1=(3q^2+1)/(4q^2), and therefore j=4(3q^2+1)^3/(q^2 g^2)=256(1-3g)^3/[g^2(1-4g)]. This is not Eq. (80), which reads 256g^2(3g-1)^3/(4g-1)^5. The discrepancy is numerical, not stylistic: at g=1/8, the correct formula gives j=8000 while Eq. (80) gives 31.25, and the paper's own Table 1 assigns j=8000 to g approximately 0.213323. Since all five g-values in Table 1 are roots of Eq. (80), the central numerical claim is unsupported. For example, the j=1728 row lists g approximately 0.198019, but solving the correct equation gives a root near g approximately 0.221, while j(0.198019) is about 2117. The conceptual suggestion that CM can occur at some couplings is not refuted by this, but the paper's specific realization is not established.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper studies the symmetric quartic Hermitian one-matrix model with potential V(x)=-x^2/2+g x^4/4 in its two-cut phase. The author derives the large-N spectral curve y^2=(x^2-a^2)(x^2-b^2), converts it to Legendre form, and computes the elliptic j-invariant j(g). Matching the resulting formula to the five positive integer CM j-values with class number one, the paper lists five admissible couplings g at which the spectral curve is claimed to have complex multiplication and enhanced automorphisms, summarized in Table 1. The advertised connection is that spectral curves of random matrix ensembles can realize complex multiplication at specific tunable couplings.","tokens_in":10373,"tokens_out":11545,"duration_ms":93303,"significance":"The conceptual goal, connecting complex multiplication to spectral curves of solvable matrix models, is attractive, and the large-N saddle-point computation leading to the spectral curve and the edges a^2=1/g+2/sqrt(g), b^2=1/g-2/sqrt(g) is standard and appears sound. If the subsequent j-invariant calculation and Table 1 were correct, the paper would provide an explicit one-parameter family of spectral curves with CM points. However, the central algebraic step from the cross-ratio to j(g) is incorrect, and every numerical entry in Table 1 is derived from that incorrect formula. The paper supplies no machine-checked derivation or reproducible code, and the quantitative claims that constitute its main message are not established as written.","major_comments":[{"comment":"Equation (80) does not follow from substituting Eq. (79) into Eq. (78). Let q=sqrt(1-4g); reading Eq. (79) as r=(-1+4g+sqrt(1-4g))/(8g-2), one obtains r=(q-1)/(2q). Substitution into the standard j-formula (78) gives r^2-r+1=(3q^2+1)/(4q^2) and r^2(r-1)^2=(q^2-1)^2/(16q^4)=g^2/q^4, hence j=4(3q^2+1)^3/(q^2 g^2)=256(1-3g)^3/[g^2(1-4g)]. This is not the expression 256g^2(3g-1)^3/(4g-1)^5 reported in Eq. (80). The discrepancy is numerical, not stylistic: at g=1/8 the correct formula gives j=8000, whereas Eq. (80) gives 31.25.","section":"Section 4, Eqs. (78)-(80)"},{"comment":"Because all five couplings in Table 1 are roots of Eq. (80), the table is invalid as a consequence of the algebraic error above. With the correct j(g), the value j=1728 is attained at g=2/9 (the minimum of the function on 0<g<1/4), not at g=0.198019 as listed, and j=8000 is attained at g=1/8, not at g=0.213323. Thus the paper's specific claim of five admissible CM couplings, and the associated automorphism-enhancement discussion in Section 4, are unsupported. The underlying idea that some CM j-values may occur for a corrected formula is not refuted, but the submitted numerical realization is wrong.","section":"Table 1 and Section 4"}],"minor_comments":[{"comment":"Equation (79) is printed without parentheses and is ambiguous; it should read r=(-1+4g+sqrt(1-4g))/(8g-2).","section":"Eq. (79)"},{"comment":"The determinant condition for the displayed SL(2,Z) matrix is written as ab-cd=1; it should be ad-bc=1.","section":"Eq. (45)"},{"comment":"Equation (64) drops the plus/minus sign in front of sqrt(b^2-4ac) and does not specify a branch choice; since tau is defined only up to PSL(2,Z) equivalence, the sign and conjugation ambiguity should be stated explicitly.","section":"Eq. (64)"},{"comment":"The symbol g is used both for the quartic coupling and for the genus (compare Eq. (30) and Eq. (38)); this notation conflict should be resolved.","section":"Sections 3-4"},{"comment":"The 'first main theorem of complex multiplication' is cited with an empty bracket before references [24, 5]; a proper reference should be supplied.","section":"Section 3.2"}],"recommendation":"reject","confidential_remarks":"The paper's main quantitative claim rests on an algebraic error in Eq. (80), and Table 1 is accordingly invalid. Although a corrected calculation might still produce some CM couplings, the submitted results do not establish the advertised realization, so I recommend rejection rather than major revision."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: the paper's one new result is the j-invariant formula (80) and the CM couplings in Table 1, and that result is wrong. Substituting Eq. (79) into Eq. (78) gives j(g)=256(1-3g)^3/[g^2(1-4g)], not 256 g^2(3g-1)^3/(4g-1)^5. I checked this by hand: with q=sqrt(1-4g), Eq. (79) is r=(q-1)/(2q); then r(r-1)=g/q^2 and r^2-r+1=(3q^2+1)/(4q^2), so the j-invariant reduces to the first expression. Numerically, at g=1/8 the correct formula gives j=8000, while Eq. (80) gives 31.25. The table's j=8000 row lists g≈0.213323, and the corrected formula is nowhere near 8000 there. Since all five rows are roots of Eq. (80), the central numerical claim is unsupported.\n\nWhat is good: the large-N two-cut solution in Sections 2-4 up to Eq. (79) is standard and clearly written. The idea of comparing the spectral curve's j-invariant with the known class-number-one list is a legitimate application, not a deep new framework but a reasonable one. There are no fitted parameters; the comparison against an external list is not circular. The references are appropriate and I see no citation-pattern problem.\n\nThe soft spot is exactly one step, but it is load-bearing: the algebraic conversion from the cross-ratio to the j-formula. The paper gives no correct algebra there, and the error propagates into Table 1, the automorphism-group remarks, and the conclusion. This is not a prefactor typo; it changes every numerical row. The general suggestion that a two-cut quartic model can have CM at some coupling may survive with the corrected formula, but this paper does not establish it.\n\nFor a reader, Sections 2-3.2 are a usable review, but the advertised result should not be cited. Worth a serious referee? No. This is a desk-reject-with-explanation, or at most a one-line referee report. I would not bring it to a reading group.","headline":"The paper's central j(g) formula is algebraically wrong: substituting its own cross-ratio gives 256(1-3g)^3/[g^2(1-4g)], not Eq. (80), so the Table 1 CM couplings are unsupported.","tokens_in":10855,"tokens_out":7604,"would_cite":false,"duration_ms":60250,"reading_group":"no","serious_thinker":"no","would_accept_peer_review":false},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["14H52","11G15","15B52"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper claims that the two-cut quartic Hermitian matrix model has a spectral curve with complex multiplication at five specific coupling values.","keywords":["complex multiplication","elliptic curves","spectral curves","Hermitian matrix models","two-cut phase","j-invariant","quartic potential","random matrix theory"],"falsifier":"A direct substitution test settles the claim: take $a^{2}$=1/g+2/\\sqrt{g}, $b^{2}$=1/g-2/\\sqrt{g}, form r=-(a-b)^2/(4ab), substitute into j=256($r^{2}$-r+1)^3/($r^{2}$(r-1)^2), and compare the simplified expression with the reported j(g)=$256g^{2}$(3g-1)^3/(4g-1)^5 at a value such as g=0.2.","tokens_in":9828,"feed_emoji":"🔢","tokens_out":10465,"duration_ms":80505,"temperature":0.7,"pith_summary":"The paper argues that in the symmetric quartic Hermitian matrix model with potential V(x)=-$x^{2}$/2+$gx^{4}$/4, the two-cut phase produces a genus-one spectral curve whose modular j-invariant is a rational function of the coupling g. By equating this j-invariant with the known integer j-values of elliptic curves with complex multiplication, the paper identifies five admissible coupling values at which the spectral curve has an endomorphism ring larger than Z. If correct, the result connects the arithmetic of elliptic curves to random matrix spectral geometry and predicts enhanced automorphism groups at those couplings.","feed_headline":"Five coupling values put complex multiplication on the spectral curve","feed_subtitle":"At five couplings the two-cut curve gains extra symmetries, linking random matrices to number theory.","key_machinery":"The load-bearing object is the spectral curve $y^{2}$=($x^{2}$-$a^{2}$)($x^{2}$-$b^{2}$) together with its cross-ratio r=-(a-b)^2/(4ab), which moves the spectral data first into the Legendre form $y^{2}$=x(x-1)(x-r) and then into a Weierstrass form. The paper's identity j(g)=$256g^{2}$(3g-1)^3/(4g-1)^5 is what converts the two-cut edge data into the modular j-invariant. The number-theoretic input is the first main theorem of complex multiplication: j is an algebraic integer for CM tori, and in the class-number-one cases the relevant j-values are the thirteen integers, five of which are positive and therefore compatible with the reported j(g).","core_discovery":"The paper's central claim is that the spectral curve of the two-cut quartic Hermitian matrix model, $y^{2}$=($x^{2}$-$a^{2}$)($x^{2}$-$b^{2}$) with $a^{2}$=1/g+2/\\sqrt{g} and $b^{2}$=1/g-2/\\sqrt{g}, has j-invariant j(g)=$256g^{2}$(3g-1)^3/(4g-1)^5 for 0<g<1/4. Since an elliptic curve with complex multiplication and class number one has integer j, the paper equates this formula to the positive integer CM j-values and obtains five admissible couplings: g approximately 0.198019 for j=1728, 0.213323 for j=8000, 0.226045 for j=54000, 0.233355 for j=287496, and 0.242923 for j=16581375. At these couplings the spectral curve acquires complex multiplication, with an enlarged endomorphism ring and, for example, an automorphism group of order four at j=1728.","pith_inferences":["The paper's search idea is modular: intersecting any closed-form j(g) with the finite list of CM j-values would generate analogous arithmetic couplings for other one-matrix or multi-matrix potentials, not only the quartic.","One testable extension is to compute higher-genus corrections at these special couplings and ask whether the free energy develops arithmetic or modular features; the paper does not pursue this.","For spectral curves of genus greater than one, the same construction would place complex multiplication on the Jacobian variety rather than on a single elliptic curve, a direction the paper indicates but leaves open."],"forward_implications":["At the five listed couplings, the spectral curve has complex multiplication, so its endomorphism ring is strictly larger than Z.","At g approximately 0.198019, the spectral curve is isomorphic to y^2=x^3-x, whose automorphism group has order four rather than two.","The remaining admissible values give spectral curves isomorphic to the explicit Weierstrass models in Table 1, so each CM point corresponds to a concrete elliptic curve in the matrix model's spectral geometry.","The construction provides a direct bridge between arithmetic properties of elliptic curves and the large-N spectral data of a random matrix ensemble."],"supporting_citations":[{"why":"Supplies the planar-diagram and resolvent saddle-point method that produces the algebraic spectral curve.","marker":"[2]"},{"why":"Provides the general random-matrix spectral-curve formalism used to set up the two-cut curve.","marker":"[12]"},{"why":"Supports the multi-cut and two-cut phase construction underlying the genus-one curve.","marker":"[10]"},{"why":"States the first main theorem of complex multiplication, that j is an algebraic integer for CM tori.","marker":"[24]"},{"why":"Supplies the complete list of class-number-one CM discriminants and integer j-values used in Table 1.","marker":"[26]"}],"fun_headline_variants":["Five couplings give spectral curves extra symmetry","Random matrices meet number theory at five couplings","Complex multiplication appears at five matrix couplings","Spectral curve gets complex multiplication at five couplings"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The claim rests entirely on the algebraic conversion of the spectral curve's cross-ratio into the reported closed form j(g); if that conversion is wrong, the five couplings and the CM conclusion at those couplings do not follow.","fun_headline_variants_meta":{"raw":{"variants":["Five couplings give spectral curves extra symmetry","Random matrices meet number theory at five couplings","Complex multiplication appears at five matrix couplings","Spectral curve gets complex multiplication at five couplings"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000525,"raw_usage":{"total_tokens":2488,"prompt_tokens":851,"completion_tokens":1637,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":467,"completion_tokens_details":{"reasoning_tokens":1583}},"tokens_in":467,"tokens_out":1637,"duration_ms":10165,"temperature":1.0,"reasoning_tokens":1583,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T15:53:15.636735+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A direct substitution test settles the claim: take $a^{2}$=1/g+2/\\sqrt{g}, $b^{2}$=1/g-2/\\sqrt{g}, form r=-(a-b)^2/(4ab), substitute into j=256($r^{2}$-r+1)^3/($r^{2}$(r-1)^2), and compare the simplified expression with the reported j(g)=$256g^{2}$(3g-1)^3/(4g-1)^5 at a value such as g=0.2.","supporting_citations":[{"cited_title":"Brézin, C","cited_arxiv_id":null,"evidence_quote":"Supplies the planar-diagram and resolvent saddle-point method that produces the algebraic spectral curve."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supports the multi-cut and two-cut phase construction underlying the genus-one curve."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"States the first main theorem of complex multiplication, that j is an algebraic integer for CM tori."},{"cited_title":"The L-functions and modular forms database","cited_arxiv_id":null,"evidence_quote":"Supplies the complete list of class-number-one CM discriminants and integer j-values used in Table 1."}],"review_version":1}