{"id":"581e4b00-096d-420e-904f-414f33be69f0","arxiv_id":"2509.06612","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A general leading-power factorization framework for exclusive multijet resolution variables, with a new rapidity regulator and an all-order factorizing k_T-ness variable.","lead":"This paper develops a general mathematical framework for simplifying multijet collision cross sections when jets are nearly unresolved, by splitting soft and collinear radiation into independent pieces. It introduces a new regularization scheme and identifies a jet-clustering variable that factorizes into simple multiplicative parts at all orders.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"All-order cumulant factorization of WTA no-recoil k_T-ness rests on an unproven monotonicity claim for WTA clustering distances; a counterexample would invalidate Eq. (4.56) even granting the Glauber-free ansatz.","rationale":"The reader's identified weak point, the Glauber-free ansatz (3.36), is real and explicitly admitted in the paper; it affects any all-order QCD claim. However, the paper qualifies its all-order statement as holding 'under the simplified assumption (3.36), i.e. neglecting Glauber effects,' so the advertised claim is already conditional on that ansatz. The more direct bridge from the kinematic analysis to the all-order cumulant factorization is the unproved monotonicity of WTA clustering distances, which the paper states without proof and which is necessary for the max property (4.54) and hence for Eq. (4.56). This is a separate, concrete, and testable assumption: if it fails, the factorization formula fails even in the Glauber-free world. The reader did not flag this, so my agreement is partial. Since both concerns are unproven but not demonstrated false, the CONDITIONAL verdict remains appropriate; the new concern reinforces it rather than changing it.","tokens_in":72283,"tokens_out":11433,"duration_ms":99730,"concrete_test":"Implement the WTA/no-recoil k_T-ness algorithm for each distance measure in Eq. (6.2), and randomly generate soft/collinear configurations with up to 8 partons assigned to sectors I1, I2, F_i, and S with the appropriate power countings. For each configuration, compute the exact leading-power k_T-ness by running the clustering sequence and record the selected distances; compare it with the sector-wise maximum in Eq. (4.54). Report any configuration where the selected distance sequence decreases or where the exact value differs from the max, with the minimal counterexample. This is a pure kinematic test requiring no higher-order QCD matrix elements.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's advertised all-order factorization for the WTA/no-recoil version of k_T-ness is established in Section 6.2 through two steps: first, a NNLO analysis showing that the leading-power approximation has the max form (4.54); second, an extrapolation to all orders based on the statement 'At leading power in the WTA scheme, the clustering distances d_1,...,d_k encountered along the clustering sequence are nondecreasing, i.e. max(d)=d_k.' No proof of this monotonicity claim is given, and it is not a general property of clustering algorithms: in other recombination schemes, merged momenta can create new distances smaller than previously selected minima. If the recorded sequence is not nondecreasing for some configuration with more than two emissions, then the exact leading-power k_T-ness would not equal the sector-wise maximum, the identity (4.56) would fail, and the advertised factorization into simple cumulant functions would not hold even under the Glauber-free ansatz (3.36). This concern is distinct from the acknowledged Glauber limitation and is directly falsifiable by a kinematic check.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript develops a method-of-regions framework for the leading-power description of exclusive multijet cross sections in the limit where a resolution variable q is much smaller than the hard scale Q. The author introduces an explicit phase-space parametrization that factorizes radiation in beam, jet, and soft sectors while tracking recoil (Sec. 4.1), combines it with a factorized ansatz for squared matrix elements (Eq. 3.36) to obtain a master factorization formula (Eq. 4.33) in terms of differential beam, jet, and soft functions, and shows that variables whose leading-power approximation is the maximum over sectors factorize into products of cumulant functions (Sec. 4.3). To treat SCET II-type variables, the author defines a rapidity regulator, the z_N-prescription, which replaces selected longitudinal momentum fractions by energy fractions, and demonstrates a zero-bin subtraction scheme that assembles all zero bins into a rapidity-finite soft-subtracted function (Sec. 5). The framework is then applied to the k_T-ness variable (Sec. 6): a region-by-region analysis at NNLO is given for E-scheme and winner-take-all (WTA) recombination, with and without beam recoil, including explicit factorization-breaking terms JSN,i and BSN. The WTA/no-recoil variant is claimed to factorize in cumulant space to all orders in perturbation theory.","tokens_in":72539,"tokens_out":24048,"duration_ms":209300,"significance":"If the framework is correct, it supplies a general slicing formalism for NNLO multijet calculations and a systematic method-of-regions alternative to SCET-based factorization derivations, together with a new rapidity regulator applicable to both SCET I and SCET II variables. The paper has substantial strengths: the recoil-aware phase-space expansion of Sec. 4.1 is carefully developed; the zero-bin hierarchy culminating in Eqs. (5.64)-(5.67) is an explicit, non-trivial construction; Appendix F provides analytic NLO cumulant jet functions for a generic class of resolution variables; and the derivations are parameter-free, with no fitting or introduced free parameters. The paper is also commendably explicit about its own limitations, including the Glauber-dependence of Eq. (3.36) and the deferral of beam and gluon z_N kernels to future work. The headline all-order claim, however, rests on an unproven monotonicity assertion about WTA clustering distances (Sec. 6.2), which is a falsifiable kinematic statement; this is the main correctness risk and should be resolved before the all-order conclusions are relied upon.","major_comments":[{"comment":"The conclusion that WTA/no-recoil k_T-ness satisfies Eq. (4.54) 'to all orders' rests entirely on the unproven assertion that the leading-power clustering distances d_1,...,d_k are nondecreasing, i.e., max(d) = d_k. This step is load-bearing: it is the only step that upgrades the NNLO analysis to the all-order cumulant factorization advertised in the abstract. The NNLO checks in Eqs. (6.31)-(6.33) cannot establish the general case, and the assertion is not a consequence of the cited literature, especially because the paper's WTA scheme deviates from Ref. [171]: Eq. (6.5) sets k_Tij = |k_ti + k_tj|, and the third distance measure in Eq. (6.2) has |k_ti + k_tj| in the denominator, so for two wide-angle soft particles the post-clustering transverse momentum can lie below the winner's transverse momentum. The claim is directly falsifiable by a kinematic scan over configurations with three or more emissions. Please either supply a proof (for instance, by arguing sector by sector that every post-clustering distance is a pre-clustering distance or a nondecreasing function of pre-clustering distances) or state the all-order claim as a conjecture and restrict the proven factorization to NNLO.","section":"Sec. 6.2, paragraph after Eq. (6.33)"},{"comment":"The abstract advertises a factorization formula for 'generic resolution variables' and a variable that factorizes 'to all orders in perturbation theory', but Sec. 3.3 states that the factorized ansatz (3.36) 'is broken at higher orders due to Glauber effects'. The all-order factorization statements of Secs. 4 and 6 therefore hold only modulo Glauber corrections, and the 'generic' claims additionally depend on the three conjectures of Sec. 2 (finite corners of the polytope R, uniform scaling within each sector, and the equivalence of single-region dominance with continuous globalness and recursive IR safety). The abstract and the theorem-level statements should carry these qualifiers; as written, the advertised conclusions are stronger than what the manuscript proves.","section":"Abstract; Sec. 3.3 and Sec. 2"},{"comment":"The z_N-prescription is constructed explicitly only for quark jet functions; the text states that modifications for beam functions and gluon jet functions are left to future work. Nevertheless, the NNLO hadron-collider k_T-ness factorization formulas (6.36) and (6.40) are written using beam and gluon-jet ingredients defined through (5.33), and the statement that the soft-subtracted functions (5.44)-(5.67) are 'already fully general' is an existence assertion about definitions that are not constructed in the manuscript. Since one advertised purpose is a slicing framework for NNLO hadron-collider processes, the paper should state explicitly which NNLO ingredients are complete and mark Eqs. (6.36) and (6.40) as provisional for gluon-initiated processes and for initial-state contributions.","section":"Sec. 5.3; Eqs. (5.32), (6.36), (6.40)"}],"minor_comments":[{"comment":"The resolution variable is typeset as 'kness_t', which reads as an unfinished placeholder; please use a properly typeset name such as k_T-ness or a dedicated symbol throughout.","section":"Throughout Section 6"},{"comment":"The plus-distribution identity for 1/z_N is the core of the scheme, but the remainder is only characterized as O(λ); a sentence stating where this remainder is dropped in the leading-power cumulant functions would improve the presentation. The 'smooth extension' from SCET I to SCET II variables in the same subsection would also benefit from a precise order-of-limits statement.","section":"Sec. 5.3, Eqs. (5.19)-(5.20)"},{"comment":"The mapping between the exponents (a,b,c) and the exponents (\\tilde a, \\tilde b) of Ref. [85] is difficult to parse; a short derivation of Eq. (2.17), or an explicit pointer, would help readers. Given that the framework's scope rests on the three conjectures in this section, a summary list of 'assumptions versus proven statements' would also improve the paper's usability.","section":"Sec. 2, Eqs. (2.13)-(2.17)"},{"comment":"The polytope R for \\sqrt{B_T \\tau} is informative, but the axes and the plotted marker coordinates are not labeled in the caption; please state them explicitly so that the two corners and the one-emission scaling point can be read off directly.","section":"Fig. 1"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is squarely within JHEP's scope, and the self-citations to Refs. [15] and [86] are substantive rather than gratuitous. The main risk is the unproven monotonicity claim behind the headline all-order factorization; if the author cannot supply a proof at revision, the claim should be downgraded to a conjecture rather than left as an assertion. A second completeness issue worth monitoring is that the beam and gluon z_N kernels remain to be constructed, which matters given active competing work on q_T-slicing with jets (Ref. [16]) and on jet-function numerics (Ref. [174])."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know. First, this is a rare paper that takes the method-of-regions approach to generic jet-resolution variables seriously and builds a factorization framework rather than just quoting SCET. Second, its advertised all-order claim for WTA k_T-ness has a load-bearing unproven step: the monotonicity of WTA clustering distances. That step is not decoration; Eq. (4.56) and the cumulant-space factorization stand on it.\n\nWhat is genuinely good: the phase-space factorization in Sec. 4.1 is unusually careful, especially the treatment of recoil and the derivation of the leading-power Born x radiation phase space. The z_N-prescription is practical and well explained; the zero-bin analysis and the construction of a soft-subtracted function are detailed enough that someone could implement them. The NLO jet functions in Appendix F are explicit. The paper is also honest about limitations: it openly states that ansatz (3.36) is broken by Glauber effects beyond NNLO, and it flags the region-identification steps as conjectures. No parameters are fitted, and the self-citations to Refs. [15] and [86] are appropriate.\n\nThe soft spots. The Glauber caveat is acknowledged, so I would not count it against the paper beyond noting that all claims beyond NNLO are conditional. The larger issue is the unproven statement in Sec. 6.2 that, at leading power in the WTA scheme, the clustering distances encountered along the clustering sequence are nondecreasing. No proof is given, and this is not a generic property of clustering algorithms. A single explicit configuration with more than two emissions where the sequence decreases would break the max identity and the all-order factorization. That is directly testable with a short kinematic calculation. It is a moderate problem because the NNLO slicing application does not need the all-order claim, but the abstract and conclusion push the all-order statement hard. I would want a proof, or the claim trimmed to something like 'factorizes to NNLO' until proven.\n\nTwo more gaps worth naming: the z_N beam functions and gluon triple-collinear kernels are deferred, so the framework is not turnkey at NNLO; and there is no numerical validation, so the finiteness of JSN,i and BSN,1 is argued, not demonstrated. Both are addressable.\n\nBottom line: read this if you work on NNLO subtraction or resummation for multijet processes. The core machinery is carefully built and the paper deserves a serious referee, but the referee should press on the monotonicity claim. If that claim breaks, the paper still stands as an NNLO framework, just not an all-order one.","headline":"A serious framework paper with an all-order k_T-ness claim that hinges on an unproven clustering monotonicity property; worth refereeing, but the referee should press on that claim.","tokens_in":72998,"tokens_out":3037,"would_cite":true,"duration_ms":30266,"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 any resolution variable that is continuously global and recursively IR safe, the leading-power exclusive n-jet cross section factorizes into beam, jet, and soft functions, and one k_T-ness variant factorizes into cumulant functions to…","keywords":["perturbative QCD","jet physics","factorization","method of regions","resolution variables","rapidity divergences","k_T-ness","NNLO slicing"],"falsifier":"Compute the first correction to the winner-take-all, no-recoil $k_T$-ness cumulant cross section from the non-factorizing soft-exchange diagrams that the paper's ansatz misses, expected beyond NNLO in the class responsible for super-leading logarithms, and check whether it can be absorbed into the product of beam, jet, and soft functions; an irreducible remainder would falsify the all-order factorization as stated.","tokens_in":72021,"feed_emoji":"⚛️","tokens_out":13633,"duration_ms":112456,"temperature":0.7,"pith_summary":"The paper develops a general method for the leading-power behavior of exclusive $n$-jet cross sections when a resolution variable $q$ is cut at $q_{\\rm cut}\\ll Q$. For any infrared-safe resolution variable that is continuously global and recursively infrared safe, the cross section is shown to factorize into hard, beam, jet, and soft functions, with the phase space expanded by the method of regions. The central payoff is a practical slicing framework: the fixed-order expansion of the factorized formula supplies the ingredients for NNLO calculations of arbitrary resolution variables, not only observables with known resummations. As an application, the paper defines a version of $k_T$-ness using winner-take-all recombination and no beam recoil whose leading-power approximation is a maximum over sectors, so the factorized convolution collapses into a product of cumulant functions to all orders. A new rapidity regularization, the $z_N$-prescription, renders the beam and jet functions finite and absorbs zero-bin contributions into a soft subtracted function suitable for numerical integration.","feed_headline":"Multijet cross sections factorize for any safe jet-resolution variable","feed_subtitle":"A winner-take-all k_T-ness splits the cross section into one cumulant function per sector, to all orders in QCD.","key_machinery":"The machinery has three moving parts. The method of regions with a power-counting polytope selects the unique contributing region per sector decomposition for admissible variables, and produces the leading-power phase space expansion, including the recoil of final-state collinear sectors against soft and initial-state-collinear radiation. The all-order factorized ansatz for squared matrix elements in mixed soft-collinear limits converts the cross section into a convolution of fully differential beam, jet, and soft functions. For variables whose leading-power limit is a maximum of sector-level pieces, the identity $\\theta(q_{\\rm cut}-\\max_i a_i)=\\prod_i \\theta(q_{\\rm cut}-a_i)$ turns that convolution into a product of cumulant functions. The $z_N$-prescription regularizes rapidity divergences by replacing selected momentum fractions with energy fractions defined through a time-like reference vector $N$, and its zero-bin contributions are absorbed into a soft subtracted function that is free of rapidity divergences and suitable for numerical evaluation.","core_discovery":"The paper's central claim is that the exclusive $n$-jet limit is governed by a single leading-power phase-space factorization rather than by observable-specific accidents. After decomposing the final state into two initial-state collinear sectors, one collinear sector per hard jet, and one soft sector, the phase space factors into a Born phase space and independent radiation phase spaces, and the squared QCD matrix element factors, under the paper's ansatz, into a hard density matrix times beam, soft, and jet kernels. For resolution variables that satisfy continuous globalness and recursive infrared safety, the approximated resolution variable in each region has homogeneous scaling, so the cumulant cross section below $q_{\\rm cut}$ is a single convolution of fully differential beam, jet, and soft functions. When the leading-power variable is a maximum of sector-level variables, the convolution becomes a product of cumulant functions; the paper shows that $k_T$-ness with winner-take-all recombination and no beam recoil has exactly this property, making its factorization exact to all orders in perturbation theory under the stated ansatz.","pith_inferences":["If the maximum-structure criterion is what drives cumulant factorization, other recoil-free, winner-take-all style observables may admit all-order cumulant-space factorization by the same argument; a natural test is to check whether a WTA variant of each existing jet-resolution variable also satisfies the max property.","Should the non-factorizing soft-exchange corrections turn out to be numerically relevant beyond NNLO, the all-order statement would need revision, but the NNLO slicing logic would survive because the factorized ansatz holds at that order; quantifying the first such correction to the WTA $k_T$-ness cumulant would map the practical range of the claim.","The failure mode of $\\sqrt{B_T\\tau}$ suggests a practical diagnostic: run the region-polytope analysis for any proposed observable, and if a fixed sector decomposition yields more than one finite corner, expect that a single product of beam, jet, and soft functions will not capture the leading-power cross section."],"forward_implications":["The factorized formula supplies ready-made beam, jet, and soft functions for NNLO slicing with any admissible resolution variable, removing the need for a dedicated all-order resummation before a slicing method can be built.","The $z_N$-prescription gives rapidity-finite beam and jet functions and a soft subtracted function that can be evaluated numerically, for both classes of resolution variables that previously required different effective-field-theory treatments.","The winner-take-all, no-recoil version of $k_T$-ness is the most convenient slicing variable for pushing beyond NNLO, because its cumulant factorization holds to all orders in perturbation theory under the paper's ansatz.","For E-scheme or recoil-collecting definitions of $k_T$-ness, the NNLO violations of the product formula are contained in the finite correction terms $J_{SN,i}$ and $B_{SN,i}$, which can be integrated numerically.","The framework explains why variables such as $\\sqrt{B_T\\tau}$ resist this treatment: their power-counting polytope has more than one corner for a fixed sector decomposition, so they are not recursively infrared safe."],"supporting_citations":[{"why":"Defines n-k_T-ness, the variable whose modified WTA/no-recoil version is shown to factorize to all orders.","marker":"[15]"},{"why":"Supplies the method of regions used to identify the phase-space regions and to derive the leading-power expansion.","marker":"[79, 80]"},{"why":"Provides the recursive-IR-safety criterion and the single-emission power-counting parametrization used to define the admissible class of resolution variables.","marker":"[85]"},{"why":"Provides continuous globalness, the condition used to ensure the power-counting polytope does not vary with the dimensionless radiation variables.","marker":"[83, 84]"},{"why":"Supplies the tree-level soft and collinear factorization results for double-unresolved limits that feed into the all-order matrix-element ansatz.","marker":"[88]"},{"why":"Supports the all-order hard-soft-collinear factorization of squared QCD amplitudes that underlies the ansatz in Eq. (3.36).","marker":"[92]"},{"why":"Introduces the time-like reference-vector regularization for rapidity divergences that the paper adapts into the z_N-prescription.","marker":"[144]"},{"why":"Defines the winner-take-all recombination scheme whose recoil-free property enables the all-order cumulant factorization.","marker":"[171]"}],"fun_headline_variants":["Exclusive n-jet cross sections factorize at leading power","Winner-take-all k_T-nes yields simple cumulant factorization","New z_N-prescription removes rapidity divergences in SCET","Generic resolution variables factorize exclusive jets","All-orders factorization for exclusive jets"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the squared QCD matrix element in simultaneous soft and collinear limits factorizes into separate soft and collinear kernels; the paper itself notes that this ansatz is broken at higher orders by Glauber effects.","fun_headline_variants_meta":{"raw":{"variants":["Exclusive n-jet cross sections factorize at leading power","Winner-take-all k_T-nes yields simple cumulant factorization","New z_N-prescription removes rapidity divergences in SCET","Generic resolution variables factorize exclusive jets","All-orders factorization for exclusive jets"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000824,"raw_usage":{"total_tokens":3640,"prompt_tokens":1018,"completion_tokens":2622,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":634,"completion_tokens_details":{"reasoning_tokens":2544}},"tokens_in":634,"tokens_out":2622,"duration_ms":18089,"temperature":1.0,"reasoning_tokens":2544,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T16:14:29.232696+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the first correction to the winner-take-all, no-recoil $k_T$-ness cumulant cross section from the non-factorizing soft-exchange diagrams that the paper's ansatz misses, expected beyond NNLO in the class responsible for super-leading logarithms, and check whether it can be absorbed into the product of beam, jet, and soft functions; an irreducible remainder would falsify the all-order factorization as stated.","supporting_citations":[],"review_version":2}