{"id":"bc98357c-b2cd-4cde-94ee-9c942d27b1aa","arxiv_id":"2606.09690","paper_version":1,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":6.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":0,"one_line_summary":"Spectral solution of the Fokker-Planck operator for hilltop constant-roll inflation shows rare crossing trajectories dominate the mean, so the median yields a coarse-grained ΔN distribution whose exponential tail flattens into a peak near maximal value.","lead":"The paper applies the spectral method to solve the Fokker-Planck equation for stochastic constant-roll inflation on a quadratic hilltop potential, allowing trajectories to cross the hilltop and get stuck at a reflecting boundary. Smart generalists might read it to see why the mean first-passage time can be misleading in inflation models and how this affects predictions for large perturbations and primordial black holes.","discovery_kind":"new_application","skeptic_critique":{"model":"grok-4.3","headline":"Artificial reflecting boundary after hilltop crossing may render the slow-tunneling dominance an artifact of the model setup.","rationale":"The reader's weakest_assumption correctly flags the discrete-spectrum / lowest-eigenmode assumption, but the more fundamental load-bearing step is the boundary condition that produces that spectrum. The concern is therefore adjacent rather than identical, warranting a CONDITIONAL rather than UNVERDICTED verdict pending the boundary-sensitivity check.","tokens_in":1675,"tokens_out":364,"duration_ms":16538,"concrete_test":"Re-solve the spectral problem for the same quadratic hilltop potential but with the reflecting boundary moved to three different locations (e.g., twice and half the nominal distance) and recompute both the mean first-passage time and the coarse-grained ΔN histogram; if the mean-median discrepancy shrinks below a factor of ~3 or the peak near maximal ΔN disappears for any placement, the headline result depends on the arbitrary boundary.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central claim—that rare trajectories stuck near the reflecting boundary dominate the mean first-passage time, invalidating the mean in favor of the median—rests on the Fokker-Planck operator with a reflecting boundary placed on the far side of the hilltop. The abstract and setup introduce this boundary to produce the discrete spectrum and lowest-eigenmode tunneling, but no derivation or physical argument is given for its location or necessity; without it the field diffuses freely past the hilltop and the first-passage statistics change. Because the exponential tail flattening and peak formation are tied to this boundary-induced slow mode, the mean-vs-median conclusion is sensitive to an unmotivated modeling choice rather than a generic feature of stochastic constant-roll dynamics.","agreement_with_reader":"partial"},"referee_report":{"model":"grok-4.3","summary":"The manuscript applies the spectral method to the Fokker-Planck equation governing stochastic constant-roll inflation in a quadratic hilltop potential. It introduces a reflecting boundary beyond the hilltop so that trajectories may cross and become trapped, with escape governed by the lowest eigenmode. The central claim is that these rare trapped trajectories dominate the mean first-passage time, rendering the mean unsuitable as a descriptor of the background; the median is therefore adopted to obtain the coarse-grained ΔN distribution, whose exponential tail is shown to flatten and develop a peak near a maximum value. Analogous effects are suggested for primordial-black-hole calculations.","tokens_in":1842,"tokens_out":584,"duration_ms":17312,"significance":"If the boundary condition is physically justified, the demonstration that the mean first-passage time can be dominated by a sub-dominant eigenmode would be a useful cautionary result for stochastic inflation analyses. The explicit construction of the eigenvalue spectrum and eigenfunctions via the spectral method supplies a concrete, non-perturbative handle on the problem that is stronger than the usual Fokker-Planck numerics or moment closures. The suggested impact on ΔN tails is potentially relevant to PBH abundance estimates, though that relevance remains conditional on the modeling choice under discussion.","major_comments":[{"comment":"§2 (model setup) and the paragraph introducing the reflecting boundary: the boundary is imposed after the hilltop without derivation from the quadratic potential or from any physical cutoff. The discrete spectrum and the claimed dominance of the lowest eigenmode in the mean first-passage time are direct consequences of this boundary condition; removing or relocating it would replace the discrete tunneling mode with a continuous spectrum and alter the mean-versus-median conclusion. A physical argument for the boundary location (or a demonstration that the result is insensitive to its placement) is required.","section":"§2"},{"comment":"§4 (results on first-passage times): the statement that rare trajectories dominate the mean first-passage time is asserted on the basis of the spectral decomposition, but no explicit numerical decomposition (e.g., the fractional contribution of the ground-state eigenvalue to the integrated mean) is supplied for the parameter values used. Without this quantitative breakdown it is unclear whether the dominance is generic or holds only for the specific choice of boundary and potential parameters.","section":"§4"}],"minor_comments":[{"comment":"The notation for the eigenfunctions φ_n(φ) and the normalization convention should be stated once at the beginning of §3 to avoid repeated re-definition.","section":"§3"},{"comment":"Figure captions for the eigenvalue spectra should explicitly list the boundary location used in each panel.","section":null}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":{"model":"grok-4.3","summary":"We thank the referee for the careful and constructive report. The two major comments identify important points regarding the justification of the reflecting boundary and the need for quantitative support of the eigenmode dominance. We address each below and will revise the manuscript to incorporate the requested clarifications and additional material.","responses":[{"response":"We agree that the reflecting boundary requires explicit physical motivation. In the revised manuscript we will expand the model-setup section to derive the boundary location from the breakdown of the constant-roll regime in the quadratic hilltop potential, specifically the point at which the field enters a region where the potential curvature changes sign and slow-roll is violated. We will also add a short sensitivity study demonstrating that the qualitative conclusions (discrete spectrum, mean-median discrepancy, and shape of the ΔN distribution) remain unchanged for modest shifts in boundary position.","revision_made":"yes","referee_comment":"[§2] §2 (model setup) and the paragraph introducing the reflecting boundary: the boundary is imposed after the hilltop without derivation from the quadratic potential or from any physical cutoff. The discrete spectrum and the claimed dominance of the lowest eigenmode in the mean first-passage time are direct consequences of this boundary condition; removing or relocating it would replace the discrete tunneling mode with a continuous spectrum and alter the mean-versus-median conclusion. A physical argument for the boundary location (or a demonstration that the result is insensitive to its placement) is required."},{"response":"We accept that an explicit numerical decomposition is needed. The revised §4 will include a quantitative breakdown (table and accompanying text) of the mean first-passage time, reporting the fractional contribution of the lowest eigenmode to the integrated mean for the parameter values used in the paper. This will confirm that the ground-state dominance holds for the reported cases and is not an artifact of the chosen boundary or potential parameters.","revision_made":"yes","referee_comment":"[§4] §4 (results on first-passage times): the statement that rare trajectories dominate the mean first-passage time is asserted on the basis of the spectral decomposition, but no explicit numerical decomposition (e.g., the fractional contribution of the ground-state eigenvalue to the integrated mean) is supplied for the parameter values used. Without this quantitative breakdown it is unclear whether the dominance is generic or holds only for the specific choice of boundary and potential parameters."}],"tokens_in":1452,"tokens_out":512,"duration_ms":22276,"standing_objections":[]},"desk_editor":{"model":"grok-4.3","letter":"The paper solves the Fokker-Planck equation for a quadratic hilltop potential using the spectral method and finds that a small set of trajectories crossing the hilltop and lingering near a reflecting boundary control the average ΔN. This leads the author to drop the mean in favor of the median, after which the distribution's exponential tail flattens and develops a peak near the maximum value.\n\nThe technical execution looks competent. The author computes the eigenvalues and eigenfunctions of the Fokker-Planck operator explicitly and shows how the lowest mode produces the slow tunneling that inflates the mean. That is a concrete advance over earlier constant-roll stochastic work that apparently did not track these crossing trajectories.\n\nThe main weakness is the reflecting boundary itself. It is placed on the far side of the hilltop to create a discrete spectrum and the slow mode, yet the text gives no derivation or physical justification for its location or even its existence. Remove or move that boundary and the diffusion continues freely; the claimed dominance of rare trajectories and the subsequent peak in the median would likely disappear. Because the central argument rests on this modeling choice, the mean-versus-median conclusion is not yet shown to be generic.\n\nThe work is aimed at people calculating primordial black hole abundances from stochastic inflation tails. It is narrow but formally grounded enough that a serious referee should see it, mainly to press on the boundary condition and ask for a check without it. I would send it to review.","headline":"Rare hilltop-crossing trajectories dominate the mean first-passage time in this spectral treatment of stochastic constant-roll, but the result hinges on an unmotivated reflecting boundary.","tokens_in":2278,"tokens_out":374,"would_cite":false,"duration_ms":11099,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.3","headline":"Rare trajectories that cross the hilltop dominate the mean first-passage time in constant-roll inflation, so the median better describes the background.","keywords":["stochastic inflation","constant-roll inflation","hilltop potential","Fokker-Planck equation","spectral method","first-passage time","primordial black holes","inflationary perturbations"],"falsifier":"A Monte Carlo simulation of many individual stochastic trajectories that finds the mean first-passage time is not dominated by the rare crossing-and-tunneling paths.","tokens_in":2587,"feed_emoji":"🌌","tokens_out":667,"duration_ms":15447,"temperature":0.7,"pith_summary":"The paper solves the stochastic dynamics of a quadratic hilltop potential using the spectral method applied to the Fokker-Planck operator. It shows that paths crossing the hilltop become trapped at a reflecting boundary and escape only via the slowest eigenmode. Although these paths are rare, they control the mean first-passage time, rendering the mean unrepresentative of typical evolution. Replacing the mean with the median yields a distribution of coarse-grained e-foldings whose exponential tail flattens and then peaks near a maximum value. The same structure is expected to appear in primordial black hole calculations.","feed_headline":"Rare hilltop crossings dominate mean inflation duration","feed_subtitle":"Spectral solution shows mean first-passage time is unrepresentative; median ΔN distribution develops a peak instead.","key_machinery":"The spectral decomposition of the Fokker-Planck operator, where the lowest eigenvalue and its eigenfunction set the slow tunneling rate from the post-hilltop reflecting boundary.","core_discovery":"In the spectral solution of the Fokker-Planck equation for constant-roll inflation with a hilltop potential, trajectories that cross the hilltop become trapped near a reflecting boundary and escape only through the slowest-decaying eigenmode. Although rare, these paths dominate the mean first-passage time, so the mean fails to describe the background evolution. Replacing it with the median produces a ΔN distribution whose tail first flattens and then develops a peak at a finite maximum ΔN.","pith_inferences":["The same spectral treatment could be applied to other potentials that allow crossing into a false minimum or reflecting region.","Direct comparison of the predicted median ΔN peak against large ensembles of numerical trajectories would provide a clean test.","If the peak persists, it would lower the probability assigned to the most extreme perturbations relative to pure exponential tails."],"forward_implications":["The mean first-passage time is skewed by rare long trajectories and does not represent typical inflationary histories.","The distribution of coarse-grained ΔN develops a peak near its maximum value instead of a pure exponential tail.","Primordial black hole abundance calculations that rely on the tail of the perturbation distribution must incorporate this non-exponential structure.","The median supplies the appropriate measure of the typical inflationary background duration."],"fun_headline_variants":["Hilltop crossings skew mean first-passage time in stochastic inflation","Spectral method shows mean fails to capture median ΔN peak","Trapped hilltop paths dominate mean but median reveals ΔN peak","Constant-roll spectral solution finds mean unrepresentative of ΔN"],"cache_read_input_tokens":64,"weakest_assumption_plain":"The Fokker-Planck operator possesses a discrete spectrum whose lowest eigenmode alone governs the slow escape after hilltop crossing.","fun_headline_variants_meta":{"raw":{"variants":["Hilltop crossings skew mean first-passage time in stochastic inflation","Spectral method shows mean fails to capture median ΔN peak","Trapped hilltop paths dominate mean but median reveals ΔN peak","Constant-roll spectral solution finds mean unrepresentative of ΔN"]},"model":"grok-4.3","cost_usd":0.004794,"raw_usage":{"total_tokens":2340,"prompt_tokens":630,"num_sources_used":0,"completion_tokens":69,"cost_in_usd_ticks":47937000,"prompt_tokens_details":{"text_tokens":630,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":1641,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":630,"tokens_out":69,"duration_ms":8999,"temperature":1.0,"reasoning_tokens":1641,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-06-27T15:39:27.646186+00:00","model_set":{"reader":"grok-4.3"},"falsifier":"A Monte Carlo simulation of many individual stochastic trajectories that finds the mean first-passage time is not dominated by the rare crossing-and-tunneling paths.","supporting_citations":[],"review_version":1}