{"id":"8352f3de-8df8-467e-9321-0993bdaad7c4","arxiv_id":"2411.10076","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":3.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Using a chaotic inflation potential with a sharp step, the paper computes PBH abundances in four mass windows and finds the GLMS approximation can make PBHs the whole of dark matter in two of them.","lead":"This paper studies an inflationary potential with a sharp step, which acts like a speed bump for the early universe's driving field and can make small, dense fluctuations collapse into primordial black holes. Using a peak-theory approximation, the authors find these black holes could account for all dark matter in two mass windows, while a simpler method gives much smaller abundances.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The f_PBH≈1 claim for the two light mass windows is driven by exp(−δ_th^2/2σδ^2) with a fixed δ_th=0.414; the paper never tests the 0.33–0.66 range it cites, and a scan could change the abundance by many orders of magnitude.","rationale":"The reader's CONDITIONAL verdict is appropriate. My stress-test did not find a reason to reject the paper's underlying machinery; standard slow-roll, Mukhanov-Sasaki, and GLMS/PS formulas provide external grounding. However, the headline f_PBH≈1 values are exponentially sensitive to the threshold and to non-Gaussianity, and the paper provides neither σδ nor a sensitivity scan. That is exactly the weakest point. The reader identified the same issue (threshold range). Thus I agree with the reader's weakest_assumption. The verdict stays CONDITIONAL: the paper should not be treated as a robust prediction of PBH dark matter until the threshold scan and non-Gaussianity check are reported.","tokens_in":15917,"tokens_out":11572,"duration_ms":124011,"concrete_test":"Repeat the abundance calculation for Set 1 and Set 2 with the P(k) used for Fig. 9: compute σδ and σ1 from Eqs. (20)–(21), then evaluate Eq. (25) for δ_th = 0.33, 0.414, 0.5, 0.66 using the same Gaussian window. If f_PBH for either light window changes by more than one order of magnitude across this range, the 'PBHs as all dark matter' claim must be reported as threshold-dependent rather than a robust model prediction. As a secondary check, estimate the skewness of the smoothed density field from the step-induced 3-point function; if |S|/σδ is non-negligible at the peak scale, the Gaussian tail formula in Eq. (25) should be replaced.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central dark-matter claim is generated almost entirely by Eq. (25), where β(M) ∝ (ν_th^2−1) exp(−ν_th^2/2) with ν_th = δ_th/σδ. Section 4.1 explicitly states that δ_th for radiation can range from 0.33 to 0.66 and that ν_th is smoothing-scale dependent, yet the paper fixes δ_th=0.414 for all four sets, reports no σδ or σ1 values, and performs no sensitivity scan. This is not a cosmetic omission: because f_PBH is exponential in ν_th^2, moving δ_th from 0.414 to 0.66 multiplies exp(−ν_th^2/2) by exp[−(1.59^2−1)ν_th^2/2]. For the ν_th≈8 implied by f_PBH≈1, that is a suppression of roughly 20 orders of magnitude; moving to 0.33 raises the abundance instead. The step-induced ultra-slow-roll episode also makes the curvature field non-Gaussian, which the manuscript itself notes can alter δ_th and the tail (Sec. 4.1), so the Gaussian integral in Eq. (25) is an unvalidated assumption at exactly the point where the headline result lives. Because Fig. 6 shows the peak amplitude changes by 10^2 under c→c+10^−7, the tuned parameter sets amplify this exponential sensitivity further. Without σδ and a threshold/non-Gaussianity scan, the values f_PBH≈1 are not a prediction but an artifact of the chosen δ_th and Gaussianity ansatz.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper studies single-field chaotic inflation with a hyperbolic-tangent sharp step in the potential. The authors solve the Mukhanov-Sasaki equation numerically, present four tuned parameter sets for which the scalar power spectrum is enhanced to O(10^-2) at small scales while the CMB-scale values of n_s and r are claimed to remain unchanged, and compute PBH abundances with the GLMS approximation of peak theory and with the Press-Schechter formalism. They report f_PBH ≈ 1 for the 10^-13 M_sun and 10^-11 M_sun mass windows and f_PBH ≈ 0.01 and 0.001 for the 1 M_sun and 6 M_sun windows, respectively. The central route—tuning a step to enhance P(k) and then applying the GLMS/PS abundance formulas—is standard, but the manuscript omits the numerical pipeline and the required sensitivity analysis, and it contains internal inconsistencies in the reported masses and CMB observables.","tokens_in":16303,"tokens_out":5109,"duration_ms":56600,"significance":"If the abundance results were robust, the model would be attractive because it decouples CMB-scale observables from small-scale PBH production and covers a wide mass range with a simple potential. The paper has several genuine strengths: it solves the full Mukhanov-Sasaki equation, explicitly studies the fine-tuning of c, d, and phi_step, and presents a side-by-side comparison of the GLMS and Press-Schechter formalisms. However, the headline f_PBH ≈ 1 claim is exponentially sensitive to the assumed density threshold and to the Gaussianity of the smoothed density field, neither of which is tested. The scientific value of the paper therefore depends on completing the robustness analysis rather than on the current numerical estimates alone.","major_comments":[{"comment":"The headline abundance values are dominated by the factor exp(-nu_th^2/2) with nu_th = delta_th/sigma_delta, but the paper fixes delta_th = 0.414 after citing a permissible range of 0.33–0.66 and reports no values for sigma_delta or sigma_1 for the four parameter sets. Because f_PBH depends exponentially on nu_th^2, moving delta_th from 0.414 to 0.66 at the nu_th values implied by f_PBH ≈ 1 suppresses the abundance by roughly twenty orders of magnitude, while moving to 0.33 raises it. Without a reported sigma_delta and a delta_th sensitivity scan, the f_PBH ≈ 1 claim is not a robust prediction and the comparison with observational constraints in Figs. 10–12 is not meaningful.","section":"Section 4.1, Eq. (25)"},{"comment":"The GLMS abundance calculation assumes the smoothed density contrast is a Gaussian random field, and the paper itself states that delta_th depends on primordial non-Gaussianities. The step-induced ultra-slow-roll phase is precisely a regime known to generate non-Gaussianity, so the Gaussian integral in Eq. (25) is an unvalidated assumption at the point where the central result lives. The authors should at least quantify the expected non-Gaussian correction to the high-density tail or show that it is negligible; without that, the agreement with observational constraints may be an artifact of the Gaussian ansatz.","section":"Section 4.1, Eqs. (22)–(25)"},{"comment":"The text states that all four parameter sets give the same n_s and r at the CMB pivot scale and that these are consistent with Planck, but Table 1 lists numerical entries for n_s and r only for set 2 (0.96 and 0.02); sets 1, 3, and 4 have empty entries. The authors should report n_s and r for every set, with uncertainties and the Planck reference values, or the claim of separate control of CMB observables is not supported by the presented data.","section":"Section 2, Table 1"},{"comment":"The conclusions contain internal inconsistencies: the text refers to PBH masses of 10^-13 M_sun and 10^-10 M_sun and to windows of 1 M_sun and 10 M_sun, while the abstract and Table 2 give 10^-13 M_sun, 10^-11 M_sun, 1 M_sun, and 6 M_sun. The manuscript also refers to a previous mass range of 10^-17 M_sun, 10^-13 M_sun, and 30 M_sun in Section 3, which is not clearly distinguished from the current results. These discrepancies must be reconciled before publication.","section":"Section 5 vs. Abstract and Table 2"},{"comment":"The paper does not describe the numerical pipeline used to solve the Mukhanov-Sasaki equation: no initial conditions, number of e-folds, k-grid resolution, Runge-Kutta step size, or convergence tests are reported. Since the peak amplitude and position of P(k) are the fundamental inputs to f_PBH and since Fig. 6 shows that the peak amplitude changes by two orders of magnitude under c -> c + 10^-7, the absence of numerical details compromises reproducibility and prevents a reader from assessing whether the reported power spectra are converged.","section":"Sections 2–3, numerical method"}],"minor_comments":[{"comment":"There is a typo: \"Press-Scheter\" should be \"Press-Schechter.\"","section":"Section 1, Introduction"},{"comment":"Several stylistic errors appear, including \"inturn,\" \"extremely light weigh,\" and \"e fold\"; a careful proofreading pass is needed.","section":"Global"},{"comment":"The notation g_s0* is confusing; the entropy degrees-of-freedom symbols should be defined consistently with the energy degrees-of-freedom g_*.","section":"Section 3, Eq. (10)"},{"comment":"The figures for f_PBH versus mass are presented with shaded observational constraints, but the numerical values of f_PBH for the four points are not given; a table of f_PBH values, with and without the sensitivity test, would be more informative.","section":"Fig. 10–12"},{"comment":"Please verify the prefactor in the GLMS expression; the standard GLMS result is often written with a factor 1/(2 pi) rather than 1/sqrt(2 pi), and the derivation leading to Eq. (25) should be shown or cited explicitly.","section":"Section 4.1, Eq. (25)"},{"comment":"The table header formatting for d/M_pl and phi_step/M_pl is not clear from the text; the units should be stated explicitly for every entry.","section":"Section 2, Table 1"}],"recommendation":"major_revision","confidential_remarks":"The paper fits the scope of a standard cosmology journal, but the f_PBH ≈ 1 result is presented as a concrete prediction without testing the two assumptions that actually generate it: the fixed threshold delta_th = 0.414 and the Gaussianity of the smoothed density field. Because both assumptions are flagged in the manuscript itself, adding the sensitivity scan and the non-Gaussianity discussion is a required, in-scope revision rather than a request for a different paper. I would not reject: the model and the GLMS/PS comparison have merit, but the numerical pipeline and the reported CMB values also need completion."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear Colleague,\n\nShort version: this is a parameter-scan paper built on a known mechanism. The sharp-step chaotic potential, the ultra-slow-roll enhancement, and the GLMS/Press-Schechter abundance formulas are all established; the same authors' 2023 paper covers the same territory. What is actually new here is four tuned parameter sets that place the scalar power spectrum peak so that PBHs form near 1e-13, 1e-11, 1, and 6 solar masses, with f_PBH from GLMS near 1 in the two light windows and 0.01/0.001 in the heavy ones. That is a modest but legitimate demonstration, and the claim that CMB-scale ns and r are unaffected by the late step is plausible and supported by the plots.\n\nThe paper does several things well. It shows the fine-tuning sensitivity of the step parameters (c changed by 1e-7 shifts the peak amplitude by two orders of magnitude), it explicitly compares GLMS and PS results, and it is transparent about the formation-threshold literature, quoting the 0.33–0.66 range rather than hiding it. It also flags one-loop backreaction as a remaining issue. This is honest work.\n\nThe soft spot is the one the stress-test note puts its finger on, and it is load-bearing. The f_PBH≈1 values are essentially determined by exp(−ν_th²/2) with ν_th = δ_th/σδ. The paper fixes δ_th = 0.414, reports no σδ values, and never scans δ_th over the 0.33–0.66 range it itself cites. Because the exponent is quadratic, that single choice changes the abundance by many orders of magnitude in either direction. The paper even notes that step-induced non-Gaussianity can alter the threshold, which makes the Gaussian integral in Eq. (25) an unvalidated assumption at exactly the point where the headline result sits. Also missing: numerical pipeline, code, convergence tests, so the P(k) curves cannot be independently reproduced. There are minor internal inconsistencies (Table 1 lists ns/r for one of four sets only; the conclusions quote 1e-10 and 10 M_sun instead of 1e-13 and 6 M_sun; Eq. (6) has a sqrt typo).\n\nOverall: the central mechanism is not in doubt, but the specific abundance claims are not yet supported. This deserves a serious referee — a desk rejection would be wrong — and the referee should ask for a threshold and non-Gaussianity sensitivity scan, σδ for each window, and the code or a full numerical pipeline before the f_PBH≈1 claim is taken at face value. If the result survives a δ_th scan, it is a useful subfield contribution; if not, it is still a clean illustration of why threshold choice dominates step-model PBH abundances.","headline":"A competent parameter-scan demonstration of the known sharp-step chaotic inflation mechanism, whose headline f_PBH≈1 is exponentially sensitive to an untested threshold choice; deserves review but needs a sensitivity scan before the dark-matter claim can be used.","tokens_in":16875,"tokens_out":3170,"would_cite":false,"duration_ms":32874,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A sharp step on a chaotic inflationary potential can form primordial black holes that could be all dark matter in two mass windows.","keywords":["primordial black holes","chaotic inflation","sharp step potential","ultra-slow roll inflation","peak theory","GLMS approximation","Press-Schechter formalism","dark matter abundance"],"falsifier":"Recompute $f_{\\rm PBH}$ for the four parameter sets using $\\delta_{\\rm th}=0.33$ and $\\delta_{\\rm th}=0.66$ (or include the non-Gaussian corrections expected from the step) and check whether the near-unity abundances for the two lightest mass windows survive; because of the exponential dependence, a fractional change in $\\nu_{\\rm th}$ of order ten percent shifts $f_{\\rm PBH}$ by orders of magnitude.","tokens_in":15686,"feed_emoji":"🕳️","tokens_out":7451,"duration_ms":64186,"temperature":0.7,"pith_summary":"This paper argues that appending a small sharp step to the standard chaotic inflation potential can amplify primordial curvature fluctuations on small scales to about $10^{-2}$, enough to form primordial black holes while leaving the cosmic microwave background observables $n_s$ and $r$ unchanged. The step acts as a speed bump, briefly slowing the inflaton field and producing a spike in the curvature power spectrum at the scales that reenter during radiation domination. Using the GLMS approximation to peak theory, the authors compute that black holes of $10^{-13}M_\\odot$ and $10^{-11}M_\\odot$ would constitute nearly all of the dark matter, while $1M_\\odot$ and $6M_\\odot$ black holes would contribute fractions of 0.01 and 0.001. A sympathetic reader would care because this is a single-field inflation model in which the physics of the CMB scale is decoupled from the physics that produces PBHs, so the model makes concrete predictions that can be checked against gravitational-wave and microlensing constraints.","feed_headline":"A speed bump in inflation forms black holes that could be all dark matter","feed_subtitle":"Chaotic inflation with a tiny step predicts black holes at four masses, two with near-total dark matter abundance.","key_machinery":"The load-bearing object is the step potential $V(\\phi)=\\frac{1}{2}m^2\\phi^2[1+c\\tanh((\\phi-\\phi_{\\rm step})/d)]$, where the dimensionless height $c$, width $d$, and position $\\phi_{\\rm step}$ are tuned to create a brief ultra-slow-roll phase that spikes the power spectrum at a chosen small scale. The abundance calculation then rests on the GLMS approximation to peak theory, specifically the analytic expression $\\beta(M_{\\rm PBH}) = \\frac{1}{\\sqrt{2\\pi}}\\left(\\frac{R\\sigma_1}{\\sqrt{3}\\sigma_0}\\right)^3(\\nu_{\\rm th}^2-1)\\exp(-\\nu_{\\rm th}^2/2)$, which replaces the full ten-dimensional peak statistics with two spectral moments $\\sigma_0=\\sigma_\\delta$ and $\\sigma_1$ of the smoothed density contrast. The smoothing is done with a Gaussian window function $W(k,R)=\\exp(-k^2R^2/2)$ on the scale $R=1/k_{\\rm PBH}$. Everything downstream depends exponentially on $\\nu_{\\rm th}^2=\\delta_{\\rm th}^2/\\sigma_0^2$, which is why the choice $\\delta_{\\rm th}=0.414$ and the Gaussian assumption for the density field carry the argument.","core_discovery":"The central claim is that the potential $V(\\phi)=\\frac{1}{2}m^2\\phi^2[1+c\\tanh((\\phi-\\phi_{\\rm step})/d)]$ with a sharp step can generate a peak in the primordial scalar power spectrum of order $10^{-2}$ at small scales, while the power spectrum on CMB scales remains nearly scale invariant with $n_s=0.96$ and $r=0.02$. Solving the Mukhanov-Sasaki equation numerically for four parameter sets, the paper finds peaks at wavenumbers corresponding to PBH masses $10^{-13}M_\\odot$, $10^{-11}M_\\odot$, $1M_\\odot$, and $6M_\\odot$. The fractional abundance is then computed with the GLMS formula $\\beta(M_{\\rm PBH}) = \\frac{1}{\\sqrt{2\\pi}}\\left(\\frac{R\\sigma_1}{\\sqrt{3}\\sigma_0}\\right)^3(\\nu_{\\rm th}^2-1)\\exp(-\\nu_{\\rm th}^2/2)$, with $\\nu_{\\rm th}=\\delta_{\\rm th}/\\sigma_0$ and $\\delta_{\\rm th}=0.414$. The result is $f_{\\rm PBH}\\approx 1$ for the two lightest windows and $0.01$, $0.001$ for the heavier ones, while the Press-Schechter formalism gives values two to three orders of magnitude smaller. The paper presents the GLMS values as consistent with current observational constraints.","pith_inferences":["If the step-induced non-Gaussianity is significant, the Gaussian GLMS estimate likely overstates the high-density tail; a calculation using the step-generated bispectrum or a numerical collapse simulation would bracket the true $f_{\\rm PBH}$.","The same small-scale power enhancement that forms PBHs necessarily sources a stochastic gravitational-wave background at second order; the paper does not compute this, but it is a testable corollary for PTA and LISA observations.","Because $f_{\\rm PBH}$ depends so steeply on $\\delta_{\\rm th}$, the 'PBH as all dark matter' conclusion should be read as contingent on the threshold and Gaussianity assumptions rather than as a robust prediction of the potential alone.","The decoupling of CMB-scale from small-scale parameters suggests the step parameters could be pinned down independently by future measurements of the scalar power spectrum at small scales, such as 21-cm observations."],"forward_implications":["The same single potential produces PBHs in four distinct mass windows, from $10^{-13}M_\\odot$ to $6M_\\odot$, by moving the step position $\\phi_{\\rm step}$.","The step leaves the CMB observables $n_s$ and $r$ unchanged, so small-scale PBH physics can be tuned independently of large-scale cosmological parameters.","In the GLMS approximation, the two lightest mass windows have $f_{\\rm PBH}\\approx 1$, meaning PBHs could constitute the entirety of the dark matter in those windows.","The heavier windows, $1M_\\odot$ and $6M_\\odot$, yield $f_{\\rm PBH}\\approx 0.01$ and $0.001$, consistent with LIGO/Virgo merger-rate constraints.","The Press-Schechter formalism underestimates $f_{\\rm PBH}$ by two to three orders of magnitude relative to the GLMS peak-theory calculation, so the choice of abundance estimator is decisive here."],"supporting_citations":[{"why":"Supplies the GLMS peak-theory approximation formula for the PBH mass fraction used in Section 4.1.","marker":"[47]"},{"why":"Provides the Press-Schechter formalism used for the comparative abundance calculation.","marker":"[48]"},{"why":"Establishes that a sharp step in a chaotic potential produces localized oscillations in the power spectrum, the mechanism this paper exploits.","marker":"[49]"},{"why":"Gives the mass-scale relation connecting PBH mass to the reentry wavenumber used in equation (11).","marker":"[79]"},{"why":"Supplies the conversion from the formation mass fraction beta to the current abundance f_PBH and the observational constraints used for comparison.","marker":"[82]"},{"why":"Supports the chosen collapse threshold delta_th = 0.414 by relating it to the equation of state.","marker":"[96]"},{"why":"Argues that delta_th depends on the shape of the collapsing overdensity, justifying the threshold sensitivity discussion.","marker":"[94]"},{"why":"Earlier study comparing peak theory and Press-Schechter abundances that the paper's two-to-three-orders-of-magnitude discrepancy claim builds on.","marker":"[84]"}],"fun_headline_variants":["Sharp step in inflation spawns black holes as dark matter","Chaotic inflation's speed bump yields near-total dark matter","Four black hole masses from a single inflation step","GLMS says black holes could be all dark matter"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The abundances assume that the smoothed density contrast is a Gaussian random field with a fixed collapse threshold $\\delta_{\\rm th}=0.414$; the result depends exponentially on this threshold and on the amplitude of the tail of the distribution.","fun_headline_variants_meta":{"raw":{"variants":["Sharp step in inflation spawns black holes as dark matter","Chaotic inflation's speed bump yields near-total dark matter","Four black hole masses from a single inflation step","GLMS says black holes could be all dark matter"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000193,"raw_usage":{"total_tokens":1444,"prompt_tokens":1132,"completion_tokens":312,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":748,"completion_tokens_details":{"reasoning_tokens":247}},"tokens_in":748,"tokens_out":312,"duration_ms":4413,"temperature":1.0,"reasoning_tokens":247,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T20:00:51.861648+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Recompute $f_{\\rm PBH}$ for the four parameter sets using $\\delta_{\\rm th}=0.33$ and $\\delta_{\\rm th}=0.66$ (or include the non-Gaussian corrections expected from the step) and check whether the near-unity abundances for the two lightest mass windows survive; because of the exponential dependence, a fractional change in $\\nu_{\\rm th}$ of order ten percent shifts $f_{\\rm PBH}$ by orders of magnitude.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the GLMS peak-theory approximation formula for the PBH mass fraction used in Section 4.1."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the Press-Schechter formalism used for the comparative abundance calculation."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes that a sharp step in a chaotic potential produces localized oscillations in the power spectrum, the mechanism this paper exploits."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the mass-scale relation connecting PBH mass to the reentry wavenumber used in equation (11)."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the conversion from the formation mass fraction beta to the current abundance f_PBH and the observational constraints used for comparison."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supports the chosen collapse threshold delta_th = 0.414 by relating it to the equation of state."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Argues that delta_th depends on the shape of the collapsing overdensity, justifying the threshold sensitivity discussion."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Earlier study comparing peak theory and Press-Schechter abundances that the paper's two-to-three-orders-of-magnitude discrepancy claim builds on."}],"review_version":1}