{"id":"0d12b440-f092-44e7-8f05-ec656b4dcce6","arxiv_id":"1908.01222","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"A heavy gluino can push the physical stop mass above LHC bounds while the underlying stop mass parameter stays near the electroweak scale, solving the little fine-tuning problem when higher-order corrections are resummed.","lead":"This paper shows that in supersymmetric models where the gluino is much heavier than the stops, the usual fine-tuning argument breaks down. By adding up corrections from all loop orders, the authors find that the physical stop mass can be large even when the underlying mass parameter is small, easing the tension with collider bounds.","discovery_kind":"first_principles","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Resummation parameter xi is not consistently defined: at the stated scale mu~mL it is positive and >1, making log(1-xi) complex; the claimed xi~-1 requires mu~M3.","rationale":"The central assertion is that resumming M3^2/mL,R^2-enhanced corrections substantially improves the fine-tuning measure. The entire resummed series is controlled by xi in Eq. (12), so the sign, magnitude, and scale dependence of xi are not peripheral details. As written, the paper contains a direct internal inconsistency: for mu~mL,R the DR-scheme xi is positive and larger than unity, making the closed-form logs complex, whereas the text's estimate xi~-1 corresponds to mu~M3. This determines the relation between DR and OS stop masses, and hence the benchmark values and Fig. 3, so a reader cannot reproduce the central numerical result from the text alone. The on-shell analysis is a meaningful independent part of the argument, and the paper deserves credit for computing 205 two-loop diagrams and for the explicit OS/DR cross-check; but that check is only as trustworthy as the definition of xi used to convert between schemes. Asking for a precise statement of the renormalization scale and a re-derivation of Eq. (12) is therefore a condition, not a rejection. This aligns with the reader's conditional verdict, although the reader's formal weakest_assumption (diagram exclusivity in the power counting) is not the concern I find most load-bearing; the xi inconsistency is directly visible in the text and can be settled by a single recalculation.","tokens_in":919,"tokens_out":956,"duration_ms":209559,"concrete_test":"Recompute xi_L from Eq. (12) for the benchmark point in Eq. (16) (M3=3 TeV, mL=611 GeV) at two scales: mu=mL and mu=M3, using alpha_s(m_t) and alpha_s(M3) respectively. If mu=mL gives xi_L>1, evaluate Eqs. (10)-(11) and check whether Im log(1-xi) is nonzero. Then recompute the fine-tuning measures Delta(mL), Delta(mR), Delta(M3) using the unambiguous OS or mu=M3 convention and compare with the values quoted in Fig. 3; if they shift by more than 20%, the resummed DR analysis is internally inconsistent.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The quantitative DR-scheme claim rests on Eq. (12), but as written the resummation parameter is not consistently defined. For the stated scale choice mu~mL,R and the benchmark value M3~5mL,R, Eq. (12) with Delta xi=0 gives xi_L approximately +2.4 (for alpha_s~0.1), not -1 as asserted in the text. With xi>1, the closed-form expressions in Eqs. (10)-(11) contain log(1-xi), which is complex, so the resummed m22^(>=3) is not a real quantity. The text's xi~-1 only follows if mu~M3, which contradicts the statement that these expressions are defined at mu~mL,R and the use of Eq. (13) to run to m_t. Since the paper claims to have numerically verified that the DR resummation and the OS shift give the same m22, this ambiguity makes the central improvement claim non-reproducible as written. The on-shell argument provides independent support and may survive a clarification, but it cannot validate the DR resummation until the scale and sign of xi are fixed.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper studies the little fine-tuning problem in the (N)MSSM when the gluino is much heavier than the stops, M3 >> mL,R. The authors argue that the standard one-loop analysis breaks down because corrections enhanced by powers of M3^2/mL,R^2 appear at every loop order. For DR-bar stop masses, they resum repeated insertions of the gluino-top stop self-energy and find that the resummed contribution to m22 is similar in size to the one-loop term, substantially improving the fine-tuning measure. They also note that in the on-shell scheme the same resummation is encoded in a large shift mL,R -> mOS_L,R, so that DR masses near the electroweak scale can coexist with on-shell stop masses above LHC bounds. The paper presents an NMSSM benchmark with Delta(mL)=6.0, an MSSM benchmark with moderate deltas, and a discussion of low-energy observables such as B-Bbar mixing.","tokens_in":9647,"tokens_out":8615,"duration_ms":88690,"significance":"If the central claim holds, the paper offers a concrete way to alleviate the little fine-tuning problem in supersymmetric spectra with a heavy gluino: the physical (on-shell) stop masses probed at colliders can be much larger than the DR-bar parameters that enter the fine-tuning analysis. The explicit two-loop calculation and the distinction between DR and OS stop masses are useful and of practical importance for low-energy flavor observables. The paper is transparent about the scheme dependence and does not claim a parameter-free prediction; the claim that LHC probes on-shell masses is a consequence of the chosen renormalization scheme, as the authors acknowledge. However, the DR resummation as presented contains a sign/scale inconsistency that must be resolved before the central improvement claim can be considered fully supported.","major_comments":[{"comment":"The resummation parameter xi is not consistently defined. For mu ~ mL,R and M3 ~ 5 mL,R, which is the stated benchmark and the stated scale of Eqs. (10)-(11), Eq. (12) with Delta xi = 0 gives xi_L,R = -(4 alpha_s/3 pi)(M3^2/m^2)[1 + log(mu^2/M3^2)]. With alpha_s ~ 0.1, the prefactor is about 1.06 and the bracket is 1 + log(1/25) = -2.22, so xi is approximately +2.35, not -1 as stated in the text. The claim xi ~ -1 is only consistent with mu ~ M3, which contradicts the sentence that the expressions define m22 at the scale mu ~ mL,R and the subsequent use of Eq. (13) to run to mt. With xi > 1, the series in Eq. (10) diverges and log(1 - xi) in Eqs. (10)-(11) becomes complex, so the resummed m22^(>=3) is not real. Consequently the numerical consistency check 'we obtain the same results for m22 in both approaches' is not reproducible as written. Please correct the sign and/or scale convention in Eq. (12), or specify the scale at which the resummation is defined, and re-run the DR/OS numerical comparison.","section":"§2, after Eq. (8)"},{"comment":"The statement that 'Other multi-loop diagrams involve fewer stop propagators and do not contribute to the highest power of M3^2/m_L,R^2' is asserted but not demonstrated. This exclusivity is load-bearing because the resummation in Eqs. (10)-(11) sums only diagrams with repeated insertions of the stop self-energy shown in Fig. 1. If other momentum-dependent or multi-gluino topologies contribute at the same parametric order, the resummed series would be incomplete and the central numerical improvement could change. Please provide a systematic power-counting argument, or an explicit bound on the omitted contributions, for the claimed M3^2/m^2 enhancement at each loop order.","section":"§2, power-counting claim"}],"minor_comments":[{"comment":"The phrase 'mass scale m1 of these new particles in in the TeV range' contains a duplicated 'in' and should read 'mass scale m1 of these new particles in the TeV range'.","section":"Abstract"},{"comment":"The definition 'g2 ≡ (g1^2 + g^2)2/2' is notationally confusing; it should presumably be (g1^2 + g2^2)/2. Please clarify the notation for the SU(2) gauge coupling.","section":"Eq. (3)"},{"comment":"The caption says 'the mean of 100 sample points' while the text says 'over 100 different parameter points'; please align the numbers. In addition, the sentence 'For most of our parameter points mOS_t1 ≈ mL' is difficult to reconcile with the benchmark in Eq. (16), where mOS_t1 = 1 TeV and mL = 611 GeV; please clarify whether this statement refers only to small-M3 samples.","section":"Fig. 3 and text around Eq. (16)"},{"comment":"The line 'Thus m2^(>=3)22 = m2^(2)22,II = 0, while m2^(2)22,I is non-zero' is missing the subscript II on the middle term; the intended statement is that only the m22II contributions vanish in the OS scheme.","section":"§2, Eq. (14)"}],"recommendation":"major_revision","confidential_remarks":"The inconsistency in Eq. (12) is the main obstacle to accepting the paper. It appears to be a sign/scale error in the definition of xi rather than a fundamental flaw, and the on-shell argument provides an independent route to the central physics, so I do not recommend rejection. However, the DR resummation is a central advertised result, and the numerical cross-check between DR resummation and the OS shift must be reproducible after the correction. I would also like to see a more explicit justification of the power-counting exclusivity claim in §2, as that assertion underpins the resummed series."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper makes a real point. In SUSY with a little hierarchy M3 ≫ mL,R, the usual fixed-order fine-tuning analysis misses corrections enhanced by powers of M3^2/m^2, and the choice between DR-bar and on-shell stop masses changes the naturalness assessment. The idea that LHC bounds on OS masses can coexist with DR masses near the electroweak scale is a genuinely new way to think about the little fine-tuning problem. The byproduct about low-energy observables probing OS masses is also sensible and useful.\n\nWhat the paper does well: It identifies a concrete failure of conventional fixed-order naturalness calculations, shows the parametric structure of the enhanced terms, and gives a clean physical explanation through the stop self-energy. The OS-scheme argument is robust because it does not rely on the problematic resummation formula. The numerical study, while based on random scans, is aimed at an honest average rather than cherry-picked minima. The example parameter points are helpful.\n\nNow the soft spots, in proportion. The stress-test concern is real. Equation (12) with μ ~ mL,R and the benchmark M3 ~ 5 mL,R gives ξ ≈ +2.4 for αs ~ 0.1, not the claimed −1. That makes log(1−ξ) complex and the resummed expressions in Eqs. (10)–(11) not real, so the DR-scheme resummation is not actually defined as written. This looks like a sign or scale mistake in the definition of ξ, not a fatal flaw in the whole idea, but it blocks verification of the central claim. The authors say they numerically checked that DR resummation and OS conversion agree, but no code or detailed derivation is provided for the summed series, so an independent referee would have to reconstruct the calculation.\n\nThe power-counting assumption that only repeated stop self-energy insertions contribute to the leading M3^2/m^2 power at each loop order is asserted rather than proven. It is plausible at the level of counting stop propagators, but I agree with the reader that this needs at least a clearer argument or a citation to a systematic expansion.\n\nWho is this for: anyone working on natural supersymmetry or on scheme dependence in effective field theory. It deserves a serious referee. The OS message may survive a clarified DR section, and the paper should not be desk-rejected. My recommendation: send it to review with a request to fix Eq. (12), define the scale of ξ precisely, and either provide the derivation or make the numerics reproducible.","headline":"A genuinely interesting scheme-dependence argument for natural SUSY that deserves a referee, but Eq. (12) as printed gives ξ≈+2.4 at μ~mL, not −1, so the DR resummation needs clarification before its central numbers can be trusted.","tokens_in":10166,"tokens_out":2054,"would_cite":true,"duration_ms":25412,"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":"The paper argues that in supersymmetric models where the gluino is several times heavier than the stops, the familiar one-loop fine-tuning analysis breaks down, while resumming the dominant loop corrections substantially improves the…","keywords":["fine-tuning","little hierarchy","supersymmetry","gluino","stop","resummation","on-shell scheme","naturalness"],"falsifier":"Compute the three-loop diagrams that are not of the repeated-self-energy form of Fig. 1, for instance diagrams with an additional independent gluino loop on a different stop line or a momentum-dependent gluino-stop vertex, and check whether any of them contributes at order $M_3^4/m_{L,R}^4$; if it does, the resummed series of Eqs. (10) and (11) is incomplete and the fine-tuning improvement could disappear. A second falsifier is experimental: a precise measurement of the on-shell stop masses showing them close to the $\\overline{\\mathrm{DR}}$ masses would contradict the large gluino-top self-energy shift $\\propto\\alpha_s M_3^2$.","tokens_in":9210,"feed_emoji":"⚛️","tokens_out":12962,"duration_ms":110671,"temperature":0.7,"pith_summary":"The paper addresses the little fine-tuning problem: heavy superpartners that couple to the two Higgs doublets inject large radiative corrections into the electroweak scale, forcing unnatural cancellations. It studies a little hierarchy in which the gluino mass $M_3$ is well above the stop mass parameters $m_{L,R}$, so the ratio $M_3^2/m_{L,R}^2$ is large. The authors show that when $M_3 \\sim 5 m_{L,R}$ the usual one-loop fine-tuning analysis is no longer trustworthy, because corrections enhanced by powers of $M_3^2/m_{L,R}^2$ appear at every loop order in the $\\overline{\\mathrm{DR}}$ scheme. Resumming the dominant chains of stop self-energy insertions tempers this growth and substantially improves the fine-tuning measure compared with fixed-order analyses. The same resummation is encoded in a shift of the stop masses from the $\\overline{\\mathrm{DR}}$ to the on-shell scheme, so the physical stop masses probed at colliders can lie well above the electroweak scale while the underlying mass parameters stay near it.","feed_headline":"Heavy gluinos cut the fine-tuning of light stops","feed_subtitle":"When gluinos outweigh stops, resummed loop corrections lower fine-tuning and push on-shell stop masses above collider bounds.","key_machinery":"The machinery is the resummation of the stop self-energy chains shown in Fig. 1: each insertion adds a gluino-top loop plus a stop mass counterterm and costs one power of $\\alpha_s M_3^2/m_{L,R}^2$. The organising parameter is $\\xi_{L,R}=-\\frac{4\\alpha_s}{3\\pi}\\frac{M_3^2}{m_{L,R}^2}\\left[1+\\log(\\mu^2/M_3^2)\\right]+\\Delta\\xi_{L,R}$, whose $\\Delta\\xi_{L,R}$ part controls the renormalisation scheme of the stop masses. In the $\\overline{\\mathrm{DR}}$ scheme the series takes the closed forms $\\sum_{k\\ge2}\\xi^k/[k(k-1)]$ and $\\sum_{k\\ge1}\\xi^k/k$, which sum to expressions involving $\\log(1-\\xi)$; in the on-shell scheme $\\Delta\\xi_{L,R}$ is chosen to cancel the self-energies, collapsing the resummation to a two-loop remnant. This object does the work of turning a fixed-order failure into a resummable geometric series, and of explaining why the on-shell and $\\overline{\\mathrm{DR}}$ stop masses can be very different.","core_discovery":"On the paper's own terms, the central claim is that the standard one-loop fine-tuning analysis of the (N)MSSM breaks down for gluino masses several times the stop masses. In the $\\overline{\\mathrm{DR}}$ scheme the two-loop contribution to the Higgs mass parameter $m_{22}^2$ contains the log-enhanced term of Eq. (9), and at higher orders the series is dominated by diagrams with repeated stop self-energies carrying the highest power of $M_3^2/m_{L,R}^2$. This series resums to the closed forms of Eqs. (10) and (11), controlled by $\\xi_{L,R}\\sim -\\frac{4\\alpha_s}{3\\pi}M_3^2/m_{L,R}^2$; for $M_3\\sim 5 m_{L,R}$ the parameter is of order $-1$, and the resummation tempers a power-like growth to a milder dependence. Switching to on-shell stop masses absorbs the resummation into a radiative shift of order $\\alpha_s M_3^2$, so the on-shell masses $m_{L,R}^{OS}$ are naturally much larger than the $\\overline{\\mathrm{DR}}$ masses. The numerical study finds moderate fine-tuning measures, e.g. $\\Delta(m_L)=6.0$, $\\Delta(m_R)=10.8$ and $\\Delta(M_3)=6.3$ for a benchmark with $M_3=3$ TeV and an on-shell stop at 1 TeV.","pith_inferences":["Beyond the paper, the same power-counting logic should apply to sbottom-like states and to any scalar whose mass is radiatively fed by a heavier fermion; a dedicated two-loop-plus calculation for the sbottom sector would test whether the improved fine-tuning survives flavour constraints.","A natural next step is to build the two-scale effective field theory with stops and gluinos integrated out at different scales; since the authors stress that their resummation differs from ordinary logarithmic RG running, a full EFT matching might expose additional power-enhanced effects at higher orders.","If this mechanism is realised in nature, a future discovery of a heavy squark would not by itself disfavour natural supersymmetry, because the physical mass could be dominated by the self-energy; telling that apart from a genuinely heavy soft mass would require precision production-cross-section and kinematic measurements."],"forward_implications":["Naturalness bounds on supersymmetric spectra weaken: $\\overline{\\mathrm{DR}}$ stop masses near the electroweak scale can coexist with on-shell stop masses above 1 TeV, because the gluino-top self-energy supplies the difference.","The fine-tuning measure drops from values of 60 or more in fixed-order MSSM scans to moderate values, with benchmark values $\\Delta(m_L)=6.0$, $\\Delta(m_R)=10.8$, and $\\Delta(M_3)=6.3$ for $M_3=3$ TeV.","Low-energy flavour observables, including $B$–$\\bar B$ mixing and rare decays such as $b\\to s\\gamma$ and $K\\to\\pi\\nu\\bar\\nu$, effectively probe the on-shell stop masses; using those masses in the leading-order prediction automatically resums the gluino-stop self-energies on the internal stop lines.","The mechanism is not tied to the details of the Higgs sector: it carries over from the MSSM to the NMSSM and to more general little hierarchies in which a heavy fermion couples to a scalar that couples to the Higgs doublets without tree-level Higgs couplings of the fermion itself."],"supporting_citations":[{"why":"Supplies the fine-tuning measure $\\Delta(p)$ used to quantify the electroweak tuning.","marker":"[10, 11]"},{"why":"Provides the recent MSSM scan with $\\Delta\\ge63$ that serves as the fixed-order baseline the paper improves on.","marker":"[42]"},{"why":"Sets the NMSSM Higgs-potential notation used for the $m_{22}^2$ parametrisation.","marker":"[43]"},{"why":"Provide the measured Higgs mass $m_h=125$ GeV that constrains the scanned parameter points.","marker":"[44, 45]"},{"why":"Generates the Feynman diagrams for the two-loop radiative corrections.","marker":"[46]"},{"why":"Supplies the (N)MSSM Feynman rules used in the diagram calculation.","marker":"[47]"},{"why":"Performs the asymptotic expansions in large masses and small external momenta that produce the two-loop results.","marker":"[48, 49]"},{"why":"Provide the analytic methods behind the asymptotic expansion of the two-loop diagrams.","marker":"[50–55]"}],"fun_headline_variants":["Heavy gluinos shrink stop fine-tuning","Resummed gluino loops cut stop fine-tuning","Little hierarchy fixes the little fine-tuning problem","Gluino mass gap eases stop fine-tuning"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument rests on the power-counting claim that at every loop order the leading correction to the Higgs mass parameter comes only from repeated insertions of the two-point stop self-energy diagram (a gluino-top loop plus a stop mass counterterm), and that no other multi-loop diagram contributes at the same order in $M_3^2/m_{L,R}^2$.","fun_headline_variants_meta":{"raw":{"variants":["Heavy gluinos shrink stop fine-tuning","Resummed gluino loops cut stop fine-tuning","Little hierarchy fixes the little fine-tuning problem","Gluino mass gap eases stop fine-tuning"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000285,"raw_usage":{"total_tokens":1817,"prompt_tokens":1221,"completion_tokens":596,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":837,"completion_tokens_details":{"reasoning_tokens":537}},"tokens_in":837,"tokens_out":596,"duration_ms":6891,"temperature":1.0,"reasoning_tokens":537,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T15:19:38.591929+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the three-loop diagrams that are not of the repeated-self-energy form of Fig. 1, for instance diagrams with an additional independent gluino loop on a different stop line or a momentum-dependent gluino-stop vertex, and check whether any of them contributes at order $M_3^4/m_{L,R}^4$; if it does, the resummed series of Eqs. (10) and (11) is incomplete and the fine-tuning improvement could disappear. A second falsifier is experimental: a precise measurement of the on-shell stop masses showing them close to the $\\overline{\\mathrm{DR}}$ masses would contradict the large gluino-top self-energy shift $\\propto\\alpha_s M_3^2$.","supporting_citations":[{"cited_title":"The current status of fine-tuning in supersymmetry","cited_arxiv_id":"1906.10706","evidence_quote":"Provides the recent MSSM scan with $\\Delta\\ge63$ that serves as the fixed-order baseline the paper improves on."},{"cited_title":"Complete set of Feynman rules for the MSSM -- ERRATUM","cited_arxiv_id":"hep-ph/9511250","evidence_quote":"Supplies the (N)MSSM Feynman rules used in the diagram calculation."}],"review_version":1}