{"id":"966fdd63-fef1-4bd3-896c-904dba6edeef","arxiv_id":"2502.07149","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The gauge origami partition function on broken lines equals a plethystic exponential and factorizes into rank-one contributions.","lead":"This paper introduces a moduli space for gauge origami on two intersecting lines, realizes it as a Quot scheme, and computes its partition function in closed form. The result gives a mathematical handle on coupled vortex systems from string theory and connects to known Nekrasov partition functions.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 2.3's zero-locus relations are internally inconsistent with Example 2.2; the intended relation is B_hat(i) I_i = 0, and the vertex computation must be redone.","rationale":"The paper's strongest asset is the closed-form plethystic exponential and the factorisation into rank-one pieces; those computations are coherent as a formal vertex computation, and the formula interpolates known smooth cases. However, none of that can define invariants of M_{r,n} if the zero-locus model is wrong. The contradiction with Example 2.2 is decisive: the paper contains two incompatible descriptions of the same moduli space, and the quiver description used for virtual classes excludes the smooth components it explicitly describes. The index error also appears in the obstruction bundle and vertex terms (Prop. 3.4, Prop. 3.5), so it cannot be dismissed as a harmless typo in one displayed equation: the computation of Corollary 3.12 depends on it. This is not a disagreement with outside consensus; it is an internal inconsistency. A modest correction of the index may rescue the formula, and the general strategy (framing independence plus scaling limits) is plausible, so the appropriate disposition is the same conditional as the reader's, pending the concrete check. I agree with the reader's weakest-assumption identification.","tokens_in":23959,"tokens_out":15911,"duration_ms":146387,"concrete_test":"Test the zero-locus statement at the point of Example 2.2: for r_1,r_2≥1 and n=1, write the representation of a quotient on Y_1 at (a,0), a≠0, as (B_1,B_2,I_1,I_2)=(a,0,1,0). Check whether it lies in Z(s) for the section of Theorem 2.3 and whether it lies in Z(s) for the corrected relations B_{\\hat i} I_i=0. Then recompute Prop. 3.4 with the corrected obstruction bundle term (replace the K_i t_i^{-1} contribution by K_i t_{\\hat i}^{-1}); if the diagonal vertex v^{(ii,αα)} and the rank-one function Z^{(1)} change, Corollary 3.12 does not follow from the corrected moduli space.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim Corollary 3.12 rests on the virtual class defined in Corollary 2.4, which in turn rests on Theorem 2.3. The relations (2.3) assert that a point in M_{r,n} satisfies [B1,B2]=0 and B_i I_i=0 for i=1,2. This is contradicted by Example 2.2: a length-1 quotient supported on Y_1 at (a,0) with a≠0 is represented by (B1,B2,I1,I2)=(a,0,1,0) (up to GL(V)), which satisfies [B1,B2]=0 and B_2 I_1=0 but B_1 I_1=a≠0. Such a point lies in M_{r,1} by the paper's own component decomposition. The correct zero-locus condition forced by the O_C-module structure is B_{\\hat i} I_i =0, since the coordinate x_{\\hat i} annihilates the i-th axis component of E_r. As stated, Z(s) excludes the smooth components Y_1,Y_2, so the virtual class of Corollary 2.4 is not the virtual class of M_{r,n}. The same index pattern enters Prop. 3.4 and Prop. 3.5 (the bundle term K_i t_i^{-1} and the diagonal vertex identity), so the vertex computation behind Corollary 3.12 is not independent of the error. The formula may survive after a correction, but the current proof does not establish it.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper introduces a moduli space of zero-dimensional quotients of a torsion sheaf on the union of two affine lines (the 'broken lines'), calls it the gauge origami moduli space M_{r,n}, and realizes it as a Quot scheme. The authors provide a quiver model and claim a global zero-locus description inside a non-commutative Quot scheme, from which they construct a virtual fundamental class and a virtual structure sheaf. They then define a K-theoretic partition function Z_r(q), compute it in closed form for all ranks via localization and a vertex formalism, and derive corollaries including a factorisation into rank-1 contributions, a Nekrasov-Okounkov twist, a cohomological limit, and relations to the Quot scheme of A^2 and to framed ADHM moduli spaces.","tokens_in":24264,"tokens_out":17831,"duration_ms":137739,"significance":"If the main theorem (Corollary 3.12) is correct, the paper provides a notable new example of a closed-form K-theoretic partition function for a singular, non-equidimensional Quot-type moduli space, with a clean factorization into rank-1 pieces and nontrivial links to Nekrasov's gauge origami and to tautological integrals on Quot schemes. The localization and vertex methods are standard, and the paper advertises parameter-free derivations and explicit formulas, which are valuable. However, the significance is conditional on the validity of the zero-locus construction, which is the foundation for the virtual class and hence for all subsequent computations.","major_comments":[{"comment":"Theorem 2.3 states that points of M_{r,n} satisfy the relations [B1,B2]=0 and B_i I_i=0 for i=1,2. This is internally inconsistent with Example 2.2. A point on the component Y_1 with support (a,0), a≠0, is represented by (B1,B2,I1,I2)=(a,0,1,0) up to GL(V); it satisfies B_2 I_1=0 but B_1 I_1=a≠0, and by Example 2.2 it lies in M_{r,1}. The same argument applies to Y_2. The correct O_C-module condition is B_{\\hat i} I_i=0, where {i,\\hat i}={1,2}. As written, the zero locus Z(s) excludes the smooth components Y_1,Y_2, so Corollary 2.4 does not define the virtual class of M_{r,n}. This is a load-bearing error: Proposition 3.4, the vertex terms in Section 3.4.1, and the localization proof of Theorem 3.11 all use the obstruction bundle associated with the section B_i I_i, not with the corrected condition.","section":"Section 2.3, Eq. (2.3)"},{"comment":"The identity v^{(ii,\\alpha\\alpha)}_n = (1 - t_i^{-1}) \\sum_{a=1}^{n_{i\\alpha}} t_{\\hat i}^{-a} is algebraically false for n_{i\\alpha} \\ge 2. Using the definition v^{(ii,\\alpha\\alpha)}_n = (1 - t_i^{-1}) Z_{n_{i\\alpha}} - (1 - t_1^{-1})(1 - t_2^{-1}) Z_{n_{i\\alpha}}^2 and Z_{n_{i\\alpha}} = \\sum_{a=0}^{n_{i\\alpha}-1} t_{\\hat i}^{-a}, a direct calculation gives (1 - t_i^{-1}) Z_{n_{i\\alpha}} t_{\\hat i}^{-n_{i\\alpha}}, not the stated sum. For example, when i=1 and n=2, the left-hand side equals (1 - t_1^{-1})(t_2^{-2}+t_2^{-3}) whereas the right-hand side is (1 - t_1^{-1})(t_2^{-1}+t_2^{-2}). This invalidates the proof of T-movability as written and casts doubt on the vertex contributions used subsequently.","section":"Proposition 3.5"},{"comment":"Because of the incorrect zero-locus relations in Theorem 2.3, the vertex term T^{vir}_n in Proposition 3.4 is not the virtual tangent space of M_{r,n}. The localization sum in Theorem 3.11 therefore does not compute the invariants of M_{r,n}. The rank-1 formula in Proposition 3.10 is independently justified via the smooth Quot scheme and is not at issue, and the factorization formula may be recoverable after a correction, but the current proof of Corollary 3.12 does not establish the closed form for the gauge origami partition function. The main claim is unproven as the manuscript stands.","section":"Corollary 3.12 and Theorem 3.11"}],"minor_comments":[{"comment":"The notation 'Hom(K_i\\cdot t_i, Q_n)' is unclear; it should specify whether the twist by t_i is on the domain W_i or on the target Q_n. The subsequent expression 'K_i t_i^{-1} Q_n' suggests a particular convention, but it is not stated consistently.","section":"Section 3.4, proof of Proposition 3.4"},{"comment":"There are several typos and small errors, e.g., 'thefore' in Section 1.3.1, 'theort' in Section 1.3.2, and 'defined as a a Nakajima' in Section 1.4. These do not affect the mathematics but should be corrected.","section":"Throughout"},{"comment":"The vanishing is stated for n>0, but the definition of Z_r(q) includes the n=0 term; the statement should specify that the coefficients for n>0 vanish in the Calabi-Yau limit, which is presumably what is meant.","section":"Corollary 3.13"}],"recommendation":"major_revision","confidential_remarks":"The central issue is not a subtle gap but a concrete inconsistency between Theorem 2.3 and Example 2.2, which undermines the virtual class construction. The intended correction is almost certainly B_{\\hat i} I_i = 0, but the vertex formalism in Section 3.4 may need to be redone with this corrected obstruction bundle, and it is not obvious that the final formula survives unchanged. The paper also contains a separate algebraic error in Proposition 3.5. I would encourage the editor to send the paper back for a careful revision rather than reject outright, because the rank-1 results and the general framework are promising, but the current proof of the main theorem is not valid."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The new object here is the Quot scheme Quot_C(E_r,n) for the union of two affine lines, with a quiver presentation, a virtual class, and a closed-form partition function for all ranks. That is a real contribution. The factorization into rank-one pieces is a nice structural result, and the links to Quot_A2 and to Nekrasov's partition function are useful. The plethystic exponential formula is clean and explicit.\n\nThe problem is Theorem 2.3. The zero locus is defined by [B1,B2]=0 and B_i I_i=0, but Example 2.2, which the paper itself uses, has points supported away from the origin on the first axis with B1 I1 nonzero. So the theorem as stated excludes components that the paper says lie in M_{r,n}. The intended relation is forced by the O_C-module structure to be B_{\\hat i} I_i=0. This is not a cosmetic typo: the section defining the zero locus has the wrong components, so the virtual class from Corollary 2.4 and the vertex weights in Proposition 3.4 inherit the error. The stress-test note is correct that the vertex computation behind Corollary 3.12 is not independent of this mistake. The formula may survive after the correction, since the corrected vertex is a natural swap of t1 and t2, but the current proof does not establish it.\n\nThere is a smaller issue in Theorem 3.8: the proof first says it will show w is non-compact and then concludes w is compact. Probably a wording slip, but it should be fixed.\n\nWhat the paper does well: the geometric setup is natural, the factorization argument is clever, and the final formula is explicit and testable. The virtual-pullback description in Section 4 gives a second route to virtual cycles that may bypass the broken Theorem 2.3; developing that route could be the cleanest repair.\n\nWho this is for: people working on Quot schemes, Donaldson-Thomas invariants, and Nekrasov partition functions. It deserves a serious referee. I would send it out, with a request that the author repair the zero-locus theorem and recompute the vertex accordingly. I would not cite it in its current form, but I would read the revision.","headline":"The new Quot-scheme moduli space and closed-form partition function are worth taking seriously, but the zero-locus theorem that carries the virtual class contradicts the paper's own Example 2.2 and must be fixed.","tokens_in":24808,"tokens_out":8868,"would_cite":false,"duration_ms":78019,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["14C05","14N35","14D21"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper proves that the K-theoretic partition function of its gauge origami moduli space on broken lines is a single plethystic exponential for all ranks, with a factorization into rank-one factors and a direct comparison to the…","keywords":["gauge origami","broken lines","Quot scheme","K-theoretic invariants","plethystic exponential","virtual fundamental class","instanton partition function","quiver representations"],"falsifier":"For $r_1=r_2=1$ and $n=1$, write out the section $s(B_1,B_2,I_1,I_2)=(e_1\\wedge e_2)\\otimes[B_1,B_2]+\\sum_i B_iI_i$ and solve $s=0$. If the printed relation $B_iI_i=0$ for $i=1,2$ is used, the component parameterising a length-one quotient on the first axis with support away from the origin disappears, contradicting the three-component description $Y_1\\cup Y_2\\cup Y_0$ of Example 2.2; replacing the relation by $B_{\\hat i}I_i=0$ restores the missing component. A direct Gr\\\"obner basis computation of the zero locus settles which relations are correct.","tokens_in":23734,"feed_emoji":"🧮","tokens_out":13894,"duration_ms":117934,"temperature":0.7,"pith_summary":"The paper introduces a moduli space of zero-dimensional quotient sheaves on the union of two affine lines, the 'broken lines' of the title, as a mathematical model for gauge origami on intersecting branes. Because the space is generally singular and has several irreducible components, the paper builds a virtual fundamental class and virtual structure sheaf through a quiver description, and defines K-theoretic invariants by equivariant localization. The main result is the closed formula $$Z_r(q)=\\operatorname{Exp}\\left(\\frac{q(1-t_1t_2)(1-$t_1^{{r_1}}$$t_2^{{r_2}}$)}{(1-t_1)(1-t_2)}\\right)$$ for all ranks, together with a factorization into rank-one factors. A sympathetic reader would care because the formula reduces an infinite sequence of virtual intersection numbers to one generating function, and links the new singular moduli space to the classical instanton partition function and to Quot schemes of the affine plane.","feed_headline":"Closed formula captures broken-line gauge origami at all ranks","feed_subtitle":"A single generating function for the K-theoretic invariants ties intersecting-line branes to instanton counts.","key_machinery":"The load-bearing mechanism is the zero-locus description of $M_{r,n}$ inside the smooth non-commutative Quot scheme: a framed quiver with one vertex, two loops, and two framing vertices produces a representation space whose quotient by $\\mathrm{GL}(V)$ carries a vector bundle whose section has zero locus exactly $\\mathrm{Quot}_C(\\mathcal{E}_r,n)$, giving the virtual class by the standard zero-locus obstruction theory. The computation is carried by the vertex term $T^{\\mathrm{vir}}_{\\mathbf n}$—the virtual $T$-representation at a torus-fixed point—whose decomposition into off-diagonal pieces makes the limiting factorization visible. The plethystic exponential is the packaging device that turns the infinite sum over fixed points into the closed rational-function formula.","core_discovery":"The paper's central claim is that the K-theoretic partition function of $M_{r,n}$ is given by the plethystic exponential above (Corollary 3.12), for every $r=(r_1,r_2)$, and that this is not an accident of low rank: the proof is a global computation valid for all $n$. The moduli space $M_{r,n}=\\mathrm{Quot}_C(\\mathcal{E}_r,n)$, with $C=Z(x_1x_2)\\subset\\mathbb{A}^2$, is cut out inside a smooth non-commutative Quot scheme as the zero locus of a section, which yields a perfect obstruction theory and virtual cycles. The torus-fixed locus is reduced and zero-dimensional, indexed by tuples of nonnegative integers; after proving framing-weight independence, the paper scales the framing parameters to infinity and obtains the factorization $$Z_r(q)=\\prod_{\\$\\alpha$=1}^{r_1}$Z^{{(1)}}$(q $t_1^{{r_1-\\alpha}}$$t_2^{{r_2}}$)\\prod_{\\$\\alpha$=1}^{r_2}$Z^{{(2)}}$(q $t_2^{{r_2-\\alpha}}$),$$ with each rank-one factor computed directly. It also shows that in the smooth case $r=(0,r)$ the virtual structure sheaf is $\\Lambda_{-t_2}T^*M_{r,n}$, recovering the equivariant $\\chi_y$-genus series, and that in general the invariants equal tautological integrals on the framed quiver moduli space of the projective plane, i.e. a classical instanton partition function with matter.","pith_inferences":["Extension: the proof structure—fixed-locus classification, framing independence, scaling to infinity—is a generally applicable recipe: any Quot-type moduli space with a proper Quot-to-Chow map and a zero-dimensional reduced fixed locus should admit an analogous factorization, although the paper states it only for broken lines.","Extension: the bubbling component $\\mathbb{P}^{r_1+r_2-1}$ in Example 2.2 suggests a moduli of expanded degenerations interpretation; establishing one would explain the three-component picture globally and might give a proper compactification of $M_{r,n}$.","Extension: testing the same machinery on a chain of $k$ affine lines would be a direct generalization; the natural conjecture, not made in the paper, is an analogous plethystic exponential with one factor per component and pairwise interaction terms.","Extension: the equality with the matter partition function of framed gauge theory gives a dictionary between the equivariant parameters $t_1,t_2$ and matter masses; a refined elliptic version would be more delicate because framing independence typically fails for higher-rank elliptic genera, as the paper notes."],"forward_implications":["For any rank pair $(r_1,r_2)$, the full series $Z_r(q)$ is known in closed form, so each coefficient can be read off by expanding the plethystic exponential without running the localization sum.","In the smooth specialization $r=(0,r)$, the virtual structure sheaf is $\\Lambda_{-t_2}T^*M_{r,n}$, so the new partition function reproduces the generating series of equivariant $\\chi_y$-genera of the Quot scheme of the affine line, providing a direct check of the virtual construction.","The broken-line invariants are equal to specific tautological integrals on the Quot scheme of $\\mathbb{A}^2$ and on the framed moduli space of the projective plane, so the result computes a classical instanton partition function with fundamental and anti-fundamental matter.","Corollaries 3.13, 3.15, and 3.16 give concrete specializations: vanishing in the Calabi-Yau limit, a square-root-twisted variant with the $[t_1t_2][t_1^{r_1}t_2^{r_2}]/[t_1][t_2]$ form, and a cohomological limit of $\\bigl((1-q)^{-1}\\bigr)^{(s_1+s_2)(r_1s_1+r_2s_2)/(s_1s_2)}$."],"supporting_citations":[{"why":"Supplies the standard perfect-obstruction-theory construction of the virtual fundamental class and virtual structure sheaf on a zero locus.","marker":"[6]"},{"why":"Provides virtual localization and Riemann-Roch in K-theory used for fixed-point computations.","marker":"[18]"},{"why":"Prior companion work whose framing-rigidity and scaling-limit argument is adapted here to prove factorization.","marker":"[20]"},{"why":"Supplies the virtual localization formula in equivariant cohomology used for the cohomological limit.","marker":"[24]"},{"why":"Provides the vertex-formalism framework for expressing the partition function as a sum over torus-fixed contributions.","marker":"[38]"},{"why":"Establishes the smoothness and dimension of the punctual Quot scheme on a line, used for the rank-one factors.","marker":"[43]"},{"why":"Defines the classical instanton partition function that the broken-line invariants are compared with.","marker":"[46]"},{"why":"Constructs the non-commutative Quot scheme and its basic properties used for the quiver model.","marker":"[64]"},{"why":"Introduces the square-root symmetrized virtual structure sheaf twist used in Corollary 3.15.","marker":"[52]"},{"why":"Provides equivariant K-theory localization, used to verify bundle identities at fixed points.","marker":"[72]"}],"fun_headline_variants":["Closed formula for broken-line gauge origami at all ranks","Broken-line gauge origami solved at all ranks by single formula","All-rank partition function unifies broken-line gauge invariants","Broken lines yield closed K-theoretic formula matching instantons"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The construction depends on the zero-locus equations (2.3) cutting out exactly the intended moduli space, but as printed those equations contradict Example 2.2 by forcing $B_iI_i=0$ on each axis and deleting the off-origin components; the intended relation is likely $B_{\\hat i}I_i=0$, and until that is corrected the virtual class is not sound.","fun_headline_variants_meta":{"raw":{"variants":["Closed formula for broken-line gauge origami at all ranks","Broken-line gauge origami solved at all ranks by single formula","All-rank partition function unifies broken-line gauge invariants","Broken lines yield closed K-theoretic formula matching instantons"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000179,"raw_usage":{"total_tokens":1328,"prompt_tokens":1003,"completion_tokens":325,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":619,"completion_tokens_details":{"reasoning_tokens":255}},"tokens_in":619,"tokens_out":325,"duration_ms":3350,"temperature":1.0,"reasoning_tokens":255,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-08T13:40:56.618080+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"For $r_1=r_2=1$ and $n=1$, write out the section $s(B_1,B_2,I_1,I_2)=(e_1\\wedge e_2)\\otimes[B_1,B_2]+\\sum_i B_iI_i$ and solve $s=0$. If the printed relation $B_iI_i=0$ for $i=1,2$ is used, the component parameterising a length-one quotient on the first axis with support away from the origin disappears, contradicting the three-component description $Y_1\\cup Y_2\\cup Y_0$ of Example 2.2; replacing the relation by $B_{\\hat i}I_i=0$ restores the missing component. A direct Gr\\\"obner basis computation of the zero locus settles which relations are correct.","supporting_citations":[{"cited_title":"Behrend and B","cited_arxiv_id":null,"evidence_quote":"Supplies the standard perfect-obstruction-theory construction of the virtual fundamental class and virtual structure sheaf on a zero locus."},{"cited_title":"Fantechi and L","cited_arxiv_id":null,"evidence_quote":"Provides virtual localization and Riemann-Roch in K-theory used for fixed-point computations."},{"cited_title":"Fasola and S","cited_arxiv_id":null,"evidence_quote":"Prior companion work whose framing-rigidity and scaling-limit argument is adapted here to prove factorization."},{"cited_title":"Graber and R","cited_arxiv_id":null,"evidence_quote":"Supplies the virtual localization formula in equivariant cohomology used for the cohomological limit."},{"cited_title":"Maulik, N","cited_arxiv_id":null,"evidence_quote":"Provides the vertex-formalism framework for expressing the partition function as a sum over torus-fixed contributions."},{"cited_title":"Monavari and A","cited_arxiv_id":null,"evidence_quote":"Establishes the smoothness and dimension of the punctual Quot scheme on a line, used for the rank-one factors."},{"cited_title":"Nekrasov, Seiberg-Witten prepotential from instanton counting , Adv","cited_arxiv_id":null,"evidence_quote":"Defines the classical instanton partition function that the broken-line invariants are compared with."},{"cited_title":"Motivic classes of noncommutative Quot schemes","cited_arxiv_id":"2303.10617","evidence_quote":"Constructs the non-commutative Quot scheme and its basic properties used for the quiver model."},{"cited_title":"Nekrasov and A","cited_arxiv_id":null,"evidence_quote":"Introduces the square-root symmetrized virtual structure sheaf twist used in Corollary 3.15."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides equivariant K-theory localization, used to verify bundle identities at fixed points."}],"review_version":1}