{"id":"73831da0-daf5-4003-8c5e-837b18cdffd8","arxiv_id":"1908.11276","paper_version":2,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"The instanton partition function of many 4d N=2 and 5d N=1 gauge theories is fully determined by the perturbative part via generalized Nakajima-Yoshioka blowup equations.","lead":"This paper presents a new way to compute instanton corrections in supersymmetric gauge theories using blowup equations, without needing to know the instanton moduli space. This matters because it opens up computations for exceptional gauge groups and exotic matter representations that were previously inaccessible.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 5d blowup equation (2.17) is conjectural; if its d-range (2.47) fails for some theory, the recursion (2.33) cannot determine Z_inst from Z_pert, so the central claim lacks proof beyond the tested examples.","rationale":"The paper presents a powerful and well-tested generalization of the Nakajima-Yoshioka blowup equations. The 4d derivation is sound: the R-charge selection rule (2.11) is rigorous, and the recursion formula (2.33) follows by a clean cancellation of the same-instanton shifted terms in (2.36). The examples are numerous, include genuinely new results (exceptional gauge groups, spinors of SO(N), rank-3 antisymmetric of SU(6)), and show agreement with ADHM, topological vertex, and superconformal index computations up to high orders. The central risk is not in the internal algebra but in the unproven 5d blowup equation (2.17). The paper itself flags this gap in Section 2.1 and again in Section 2.3, where the range (2.47) is called a conjecture based on an empirical pattern. This is exactly the load-bearing assumption: if (2.17) fails for some theory, the recursion no longer determines Z_inst from Z_pert, and the headline claim would be false for that theory. The Zextra issue is secondary but real: in several cases the recursion reproduces a partition function with a spurious factor, so the phrase \"instanton partition function\" is not always the bare QFT observable. Both issues are disclosed, and the evidence makes the conjecture plausible, but they prevent a full ACCEPT. A CONDITIONAL verdict is appropriate, pending a proof or at least a first-principles derivation of the 5d d-range, or a decisive boundary-case test. The reader identified the same weakest assumption, so the verdict should remain unchanged.","tokens_in":50954,"tokens_out":15282,"duration_ms":138907,"concrete_test":"Compute the 1- and 2-instanton partition functions for 5d SU(6) with one hypermultiplet in the symmetric representation (a UV-complete theory with dmax=2, not covered in Table 1) using an independent method, e.g., the topological vertex on a 5-brane web or a superconformal index computation. Then check whether the recursion formula (2.33) with d=0,1,2 and the universal 1-instanton formula (2.39) reproduce those results. Exact agreement would support the conjectured sufficiency of (2.47) at a boundary case; any mismatch would refute it and invalidate the central claim for that theory.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that the instanton partition function is completely determined by the perturbative part rests on the blowup equation (2.17) holding for at least three distinct values of d. In 4d this follows from an R-charge selection rule, but in 5d the equation is not derived. The authors state explicitly (Section 2.1): \"We conjecture that the bound on d we obtain is actually sufficient to obtain the blowup equation (2.17). While we do not attempt to prove this sufficiency.\" The allowed d-range (2.47) is inferred from the empirical (s,s') pattern (2.46), obtained by expanding the one-instanton equation (2.42) in powers of p1p2 and requiring consistency (2.43)-(2.45). This is at best a necessary condition at one-instanton order, not a proof of (2.17) at all instanton orders. If the conjectured range is too optimistic for some theory—for example, a representation with dmax=2 exactly, or a theory outside the tested list—then fewer than three independent equations hold, and Z_inst is no longer determined by Z_pert. The paper's extensive checks against ADHM, topological vertex, and superconformal index make the conjecture plausible, but they do not close this logical gap. Additionally, in cases with Zextra≠1 (e.g., SU(N) with Nf+2|κ|=2N, Section 3.1), the recursion reproduces a string-embedded partition function containing a spurious factor; the unambiguous QFT observable requires outside input to remove Zextra. This reinforces that the unqualified phrase \"the instanton partition function\" in the abstract is stronger than what is actually established.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a generalisation of the Nakajima-Yoshioka blowup equations to arbitrary gauge groups and hypermultiplet representations, and uses these equations to derive a recursion relation (2.33) for the instanton partition function together with a closed one-instanton formula (2.39). The formalism is applied to a large class of 4d N=2 and 5d N=1 theories, including exceptional gauge groups, SO(N) theories with spinor matter, and SU(6) with a rank-3 antisymmetric hypermultiplet, and it is tested against ADHM localisation, topological vertex computations, and superconformal index results. The central claim, stated in the abstract and in Section 2.2, is that the instanton partition function is completely determined by the perturbative part.","tokens_in":51254,"tokens_out":6281,"duration_ms":61409,"significance":"If the underlying 5d conjecture is correct, the result is significant: it bypasses the need for an explicit instanton moduli space construction, produces previously unknown partition functions for exceptional and exotic matter theories, and gives a strikingly strong relation between perturbative and non-perturbative data. The paper's strengths include the clean 4d derivation from the U(1)_R selection rule, the detailed and extensive comparisons with independent methods, the explicit character expansions given in Appendix A, and the nontrivial Higgsing check for the SU(6)+TAS theory. The significance is conditional, however, because the 5d blowup equation is not derived and the determinacy claim is also qualified by the appearance of spurious Z_extra factors in several classes of examples.","major_comments":[{"comment":"The 5d blowup equation (2.17) is the load-bearing input of the recursion (2.33), but it is not derived. The 4d selection-rule argument of Section 2.1 has no analogue in 5d, and the text states: \"We conjecture that the bound on d we obtain is actually sufficient to obtain the blowup equation (2.17). While we do not attempt to prove this sufficiency.\" The allowed range (2.47) is obtained as a necessary condition by expanding the one-instanton equation (2.42) and requiring consistency at leading order, which yields (2.45); the empirical (s,s') pattern (2.46) is then extrapolated to all instanton orders. A necessary condition at one-instanton order does not establish (2.17) at arbitrary instanton number. If for some theory fewer than three d-values satisfy (2.17), for example a theory with d_max=2 or a representation outside the tested list, the system (2.34)-(2.36) is underdetermined and the central claim that Z_inst is completely determined by Z_pert fails. The paper should either prove the sufficiency of the conjectured range or explicitly present the determinacy statement as a conjecture supported by the listed examples, and adjust the abstract accordingly.","section":"§2.1 and §2.3, Eqs. (2.17), (2.46)-(2.47)"},{"comment":"For several classes, the recursion reproduces an ADHM partition function that contains an extra factor Z_extra. As the authors explain, this factor is independent of the Coulomb VEV and is spurious from the 5d QFT perspective; it appears when N_f+2|κ|=2N for SU(N), when N_f+2|κ|=N+4 for SU(N)+AS, and when N_v=N-4 for SO(N). Removing Z_extra requires outside information, namely a choice of string-theory embedding. This directly qualifies the abstract claim: the perturbative part alone does not determine the QFT observable Z_inst; it determines a string-embedded partition function only up to a VEV-independent factor. The paper should state this qualification wherever the determinacy claim is made, rather than only in the comparison sections.","section":"§3.1, Eqs. (3.1)-(3.2)"},{"comment":"The one-instanton formula (2.39) is claimed to be independent of the choice of d_0 and to hold universally for any gauge theory with d_max>2. This independence is not proved, and Section 3.1 shows that for SU(2)+N_fF with N_f≥5 the formula disagrees with the correct Witten index of the D0-D4-D8-O8 system, agreeing instead with a colliding-brane web partition function. Thus the word \"universal\" is too strong as stated. The formula should be presented as valid under the conjectural blowup-equation range, with the SU(2), N_f≥5 exception stated explicitly at the point where (2.39) is introduced.","section":"§2.2 and §2.3, Eq. (2.39)"}],"minor_comments":[{"comment":"The recursion formula (2.36) divides by (1-p_1^n)(1-p_2^n); please clarify the precise sense in which the identity holds at resonant values p_1^n=1 or p_2^n=1, given that the Nekrasov partition function is meromorphic and the recursion is used as a formal series identity.","section":"§2.2, Eq. (2.36)"},{"comment":"The comparisons with [71] are reported as agreements 'up to t^20 order' or 'up to t^280 order' without specifying the computational method or the truncation used for the character reductions. A brief statement of the verification procedure would make these tests more reproducible.","section":"§3.2, text after Eq. (3.10)"},{"comment":"The topological vertex computation in Section 3.4 is performed in the unrefined limit 2ϵ_+=0, while the blowup recursion is formulated in the refined case. Please state explicitly whether the refined comparison was also checked for Z_1 and Z_2, or whether only the unrefined limit was verified.","section":"§3.4, Eqs. (3.34)-(3.47)"},{"comment":"The symbols N_f, N_s, N_c, N_v are used both as counts of hypermultiplets and as subscripts on characters in Appendix A; a short glossary of this notation near Table 1 would improve readability.","section":"Notation throughout"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is transparent about its main conjecture, and the volume of independent cross-checks makes the computational claims credible. My main concern is that the abstract and central statements assert a stronger determinacy result than the paper actually proves, especially in 5d and in the presence of Z_extra. I would not reject the paper; I would ask for the claims to be qualified to match the proven and conjectural parts, and for the sufficiency of the d-range to be either proved or explicitly kept as an assumption throughout."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Glad you asked about this one. Short version: it deserves a serious referee. The paper does something real: it takes the Nakajima-Yoshioka blowup machinery and extends it to arbitrary gauge groups and arbitrary matter representations, then uses it to produce instanton partition functions that no one had computed, including exceptional groups with fundamentals, SO(N) spinors, and SU(6) with a rank-3 antisymmetric hypermultiplet. The one-instanton closed form (2.39) is genuinely useful and depends only on group-theoretic data. The checks are substantial: ADHM agreement for SU(N), Sp(N), SO(N); matches with the superconformal index/topological vertex results for exceptional and spinor cases; and extremely high-order agreement for E8 and E7. The 4d derivation from the U(1)_R selection rule is clean and convincing. The citation pattern is honest, crediting Keller-Song and Del Zotto-Lockhart where due and flagging where the paper extends or corrects them.\n\nThe soft spot is exactly where the authors say it is. The 5d blowup equation (2.17) is not derived; it is conjectured to hold for a range of d that is inferred from an empirical (s,s') pattern coming from the one-instanton equation. That pattern is a necessary condition at one-instanton order, not a proof at all instanton orders. The recursion (2.33) needs at least three independent d values; if the conjectured range fails for some theory, the 'perturbative part determines everything' claim collapses. The authors say this plainly in Section 2.1, and the extensive checks make the conjecture plausible, but they do not close the gap. I think the abstract oversells slightly when it says the instanton partition function is completely determined by the perturbative part, because in the Zextra≠1 cases the recursion reproduces a string-embedded partition function with a spurious factor, not the unambiguous QFT observable. Again, this is disclosed in Section 3.1, but it is a real qualification.\n\nWho is this for? Anyone computing Nekrasov partition functions for theories without ADHM, and anyone interested in blowup equations for topological strings. It is not a finished theorem, but it is a well-tested proposal with open problems clearly stated. If I were the editor, I would send it to a knowledgeable referee — the math is checkable and the stakes are high. A revision that either proves the d-range bound or derives it from a physical principle, and that deals with the Zextra ambiguity, would make the paper close to definitive. As is, it deserves peer review, and I would cite the one-instanton formula.","headline":"A strong, honest paper that computes many new instanton partition functions via generalized blowup equations, with the main caveat being the explicitly conjectural 5d validity bound.","tokens_in":51855,"tokens_out":5476,"would_cite":true,"duration_ms":43427,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"For a large class of 4d and 5d gauge theories, the full instanton partition function follows from the perturbative partition function through blowup equations.","keywords":["blowup equations","Nekrasov partition function","instanton counting","5d N=1 gauge theory","4d N=2 gauge theory","exceptional gauge groups","spinor matter","rank-3 antisymmetric tensor"],"falsifier":"Compare the three-instanton term from the recursion (2.33) for $SO(10)$ with two spinor hypermultiplets against a topological vertex computation from the 5-brane web; a mismatch at generic Coulomb VEV would falsify the conjectured validity range.","tokens_in":50683,"feed_emoji":"🧮","tokens_out":6172,"duration_ms":59141,"temperature":0.7,"pith_summary":"The paper claims that for any 4d $\\mathcal{N}=2$ or 5d $\\mathcal{N}=1$ gauge theory whose matter content is not too large, the full instanton partition function is recovered from the perturbative (one-loop) data alone. The mechanism is a set of blowup equations: localizing the theory on $\\mathbb{C}^2$ blown up at a point produces relations between shifted partition functions, and when enough independent relations exist, each $n$-instanton term is determined recursively. This removes the need to construct the instanton moduli space explicitly, which was previously an obstacle for exceptional gauge groups and exotic matter representations. If correct, instanton counting becomes a group-theoretic exercise for a wide class of theories, including cases where no ADHM construction exists.","feed_headline":"Instanton terms are fixed by perturbative data alone","feed_subtitle":"Blowup recursion determines all instanton corrections for exceptional groups and exotic matter without ADHM constructions.","key_machinery":"The central object is the one-point blowup equation. On the blowup $\\widehat{\\mathbb{C}}^2$, localization expresses the full partition function as a flux sum over products of two flat-space partition functions with shifted Coulomb, mass, and fugacity parameters. Inserting powers of a topological operator associated with the blown-up two-cycle produces a family of identities labelled by an integer $d$; when the matter content leaves at least three allowed values of $d$, the equations become linear relations that determine the unknown $n$-instanton term. The universal one-instanton formula is a sum over long roots $\\Delta_{\\ell}$ with weight-vector contributions from each hypermultiplet, so the output is built entirely from group-theoretic data.","core_discovery":"The central claim is that the Nekrasov partition function obeys a blowup identity $Z = \\sum_k Z^{(N),d}(k) Z^{(S),d}(k)$ for $0 \\le d \\le d_{\\max}$ whenever the matter representation is not too large, and that three independent such equations suffice to solve recursively for every instanton term. In particular, the one-instanton partition function has the universal closed form given in equation (2.39), expressed in terms of the long roots, weights, and Dynkin indices of the gauge group and matter representations; higher instanton orders follow from the recursion (2.33). The paper's stated conclusion is that the instanton partition function is completely determined by the perturbative part. The precise validity range for 5d theories is partly conjectural, based on an empirical exponent pattern rather than a proof.","pith_inferences":["An unresolved but natural next step is to derive the 5d validity range (2.47) from a first-principles selection rule analogous to the 4d R-charge argument; if found, it would place the conjectural range on the same footing as the 4d case.","The empirical pattern that $s=s'=h^\\vee-\\frac12\\sum I_2(R_l)$ for all non-SU(N) groups suggests a universal bound in the absence of Chern-Simons and theta terms; testing it on additional representations would be a direct check.","The existence of such recursion suggests that similar blowup identities may hold for other equivariant partition functions, such as indices on orbifolds or elliptic genera, potentially yielding new constraints on 6d theories."],"forward_implications":["Instanton partition functions for exceptional gauge groups and exotic matter representations become computable without an ADHM construction, directly from perturbative one-loop data.","The recursion determines all higher-instanton terms once the one-instanton term is known, so instanton counting reduces to evaluating group-theoretic sums.","For theories with a spurious extra factor, the blowup equations select a particular string theory embedding, explaining discrepancies between different UV completions at the level of the partition function.","The same logic applies to 4d theories, where the selection rule is rigid, giving a general route to 4d instanton partition functions."],"supporting_citations":[{"why":"Introduces the blowup decomposition of the instanton partition function on the one-point blowup, the foundational identity generalized here.","marker":"[5]"},{"why":"Establishes the K-theoretic blowup equation for pure SU(N), the 5d starting point extended to arbitrary gauge groups and matter.","marker":"[21]"},{"why":"Derives the recursion formula for instanton counting from blowup equations, which the paper adapts to include hypermultiplets.","marker":"[22]"},{"why":"Uses the blowup formula to compute exceptional instantons without matter, demonstrating the method that this paper generalizes.","marker":"[28]"},{"why":"Provides the JK-residue localization results for ADHM quantum mechanics used to test the recursion against known instanton partition functions.","marker":"[15]"},{"why":"Gives supersymmetric quantum mechanics and Young-diagram expressions for exceptional instantons used as a comparison benchmark.","marker":"[66]"},{"why":"Supplies known one-instanton partition functions for theories with spinors and exceptional groups, used to validate the universal one-instanton formula.","marker":"[71]"},{"why":"Constructs the 5-brane web for SU(6) with a rank-3 antisymmetric tensor and the Higgsing check used to test the two-instanton result.","marker":"[55]"}],"fun_headline_variants":["Blowup recursion fixes all instanton terms","No ADHM needed: blowup equations fix instantons","Instanton terms come solely from perturbative data","Exceptional gauge groups: blowup solves instantons"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole recursion rests on the conjecture that the blowup equations hold for the full range of $d$ in (2.47), which is inferred from a one-instanton exponent pattern rather than proved.","fun_headline_variants_meta":{"raw":{"variants":["Blowup recursion fixes all instanton terms","No ADHM needed: blowup equations fix instantons","Instanton terms come solely from perturbative data","Exceptional gauge groups: blowup solves instantons"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001152,"raw_usage":{"total_tokens":4707,"prompt_tokens":807,"completion_tokens":3900,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":423,"completion_tokens_details":{"reasoning_tokens":3838}},"tokens_in":423,"tokens_out":3900,"duration_ms":26517,"temperature":1.0,"reasoning_tokens":3838,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T10:19:48.801735+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compare the three-instanton term from the recursion (2.33) for $SO(10)$ with two spinor hypermultiplets against a topological vertex computation from the 5-brane web; a mismatch at generic Coulomb VEV would falsify the conjectured validity range.","supporting_citations":[{"cited_title":"Counting Exceptional Instantons","cited_arxiv_id":"1205.4722","evidence_quote":"Uses the blowup formula to compute exceptional instantons without matter, demonstrating the method that this paper generalizes."},{"cited_title":"6d strings and exceptional instantons","cited_arxiv_id":"1801.03579","evidence_quote":"Gives supersymmetric quantum mechanics and Young-diagram expressions for exceptional instantons used as a comparison benchmark."}],"review_version":1}