{"id":"71a25396-aa5b-4aa2-8ff0-eca9f3b4c31c","arxiv_id":"2608.10105","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Paleo-detectors are projected to reconstruct WIMP masses from 1 GeV/c² to 1 TeV/c² and to reject canonical spin-independent or spin-dependent interaction hypotheses for most non-standard NREFT operators without recoil direction information.","lead":"This paper projects how well ancient minerals used as paleo-detectors could reconstruct the mass of dark matter particles and tell apart different dark matter nucleus interaction types. It finds these detectors could measure light WIMP masses near 10 GeV and distinguish standard from non-standard interactions without needing directional information.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"All mass-reconstruction and operator-discrimination claims flow through the deterministic Eq. (7) mapping from recoil energy to track length; Sec.","rationale":"The paper is a forecasting study, and the headline capabilities are shape-based: mass reconstruction and operator discrimination both rely on differences in the binned track-length spectrum. The link between the physics (recoil energy spectrum) and the observable (track lengths) is Eq. (7), a single-valued SRIM range. The authors themselves flag in Sec. VII that this mapping is not valid at sufficiently low masses, where a given recoil energy maps to a distribution of track lengths. The low-mass projection (m_chi <~ 10 GeV, Delta m/M ~ 0.1) and the short-track discrimination regime (m_chi <~ 30 GeV in the LR scenario) are exactly the places where track lengths are short and range straggling and track-formation stochasticity should be largest. If the spread in track length is comparable to the bin size or to the separation between operator spectra, both the reconstructed mass intervals and the operator-discrimination p-values shrink. The reference to Ref. [24] gives a reason to think the approximation is acceptable in the 1 GeV-5 TeV range, but the paper does not show the residual smearing or repeat any of its own analysis with a distribution. Because no code or data are released, this cannot be checked from the manuscript. The reader's conditional verdict is therefore appropriate, and the single test that would settle the issue is to redo the Asimov profile-likelihood analysis for representative benchmarks using a track-length distribution. Other concerns (Wilks coverage, factor-of-2 comparison, O8/O7 exceptions) are real but secondary: they affect the size or phrasing of the claims, not the underlying observable mapping.","tokens_in":35579,"tokens_out":8731,"duration_ms":85594,"concrete_test":"Recompute the two benchmark projections with a track-length distribution instead of Eq. (7): use p(x_T|E_R) from Ref. [24] if available, or a Gaussian with width sigma_x(E_R) = 0.1 x_T(E_R) (SRIM-style range straggling) for gypsum and halite. Specifically, redo the Asimov profile-likelihood mass-interval calculation for m_chi = 5 GeV/c^2 in the HR scenario and m_chi = 500 GeV/c^2 in the HE scenario in Fig. 3, and the operator-rejection p-values in Figs. 5 and 7. If the 2-sigma mass intervals broaden by more than a factor ~2, or if the exposure required for 3-sigma rejection of the standard SI/SD hypothesis shifts by more than a factor ~2 relative to the deterministic-mapping results, the paper's central claims must be restated as conditional on the track-length model.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central projections all pass through the track-length spectrum as a proxy for recoil energy. Equation (7) assumes a deterministic, one-to-one mapping x_T(E_R) computed from SRIM stopping powers. The authors concede in Sec. VII that for low masses a recoil with a given E_R produces a distribution of track lengths and that the deterministic mapping 'is no longer valid' in that regime. The headline low-mass reconstruction claim (m_chi = 1-10 GeV/c^2, Delta m_chi/M_chi ~ 10^-1) and the short-track discrimination sensitivity near m_chi ~ 10-30 GeV/c^2 are precisely the regimes where tracks are only a few nm to tens of nm and range straggling and track-formation stochasticity are largest. If the width of x_T for fixed E_R is comparable to the bin spacing or to the spectral separation between operators, the likelihood-ratio separation that drives both capabilities is reduced. Ref. [24] is cited for the accuracy of the deterministic mapping over 1 GeV/c^2 to 5 TeV/c^2, but the paper does not quantify the residual distribution or demonstrate that the projected mass intervals and p-values are robust to it. Since no code or data accompany the paper, this assumption cannot be checked from the manuscript. It is the most load-bearing because every reconstruction and discrimination contour inherits the uncertainty in the energy-to-track-length relation.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper presents a likelihood-based forecasting study for paleo-detectors. Assuming that a WIMP signal has been detected in the track-length spectrum of an ancient mineral (gypsum in the main text, halite in the appendix), the authors use Asimov datasets and profile log-likelihood ratios to project (i) how well the WIMP mass can be reconstructed for a set of NREFT operators in both elastic and inelastic scattering, and (ii) how confidently the standard spin-independent (O1s) or spin-dependent (O4s) hypothesis can be rejected when the true signal is generated by a non-standard operator. The headline results are that masses below about 10 GeV/c2 could be reconstructed with relative uncertainties around 0.1 in the high-resolution scenario, masses up to about 1 TeV/c2 could be reconstructed in the high-exposure scenario, and many non-standard operators could be distinguished from canonical ones without directional information at higher masses. The analysis is explicitly conditional on single-operator dominance, isoscalar couplings, and a fixed total signal rate, and these limitations are stated in the text.","tokens_in":35737,"tokens_out":7970,"duration_ms":83033,"significance":"If the projections hold, this is a useful and timely complementarity study: it extends paleo-detector forecasts from sensitivity limits to parameter reconstruction and operator discrimination, covers a broader operator set than previous work, and treats elastic and inelastic scattering on a common footing. The statistical framework is standard and applied consistently, and the paper is transparent about the main assumptions, including the operator-conditional nature of the mass reconstruction and the fixed-normalization, shape-only information content of the likelihood. The main reservation is that the reach of all forecasts is inherited from the deterministic track-length model of Eq. (7), and the manuscript does not yet quantify how robust the headline projections are to the breakdown of that model at short track lengths.","major_comments":[{"comment":"The headline low-mass reconstruction (Fig. 3, Mχ = 1–10 GeV/c2) and the short-track discrimination curves (Fig. 5 left, mχ = 5 GeV/c2) rely on spectral differences at track lengths of only a few to tens of nanometers. Equation (7) assumes a deterministic, one-to-one mapping x_T(E_R) computed from SRIM stopping powers, and the paper cites Ref. [24] for the accuracy of this approximation. However, Sec. VII itself states that at lower masses a recoil with fixed E_R yields a distribution of track lengths and that the deterministic mapping is no longer valid in that regime. The manuscript should quantify the width of this track-length distribution for gypsum and halite and demonstrate that the projected Δmχ/Mχ values and p-values are stable under a conservative smearing of the track-length spectra; without such a test, the low-mass portion of the central claim is not yet fully supported.","section":"Sec. III, Eq. (7); Sec. VII"},{"comment":"The abstract and introduction state that for WIMP masses ≳ 10 GeV/c2 the canonical SI/SD hypotheses can be excluded for nearly all non-standard operators without directionality. The results in Sec. VI and Fig. 7 show a more nuanced picture: in the low-resolution scenario that the authors themselves adopt for mχ > 10 GeV/c2, discrimination power is significantly reduced below mχ ≈ 20–30 GeV/c2 (and below about 40 GeV/c2 for the inelastic case), and the operators O8s (and O7s in halite) are never discriminable; furthermore, Fig. 5 left shows no LR-scenario rejection at 5 GeV/c2. The abstract should be reworded to match the paper's own conclusion, which uses the more accurate phrase \"above a few tens of GeV,\" and should state the exceptions explicitly.","section":"Abstract; Sec. VI, Figs. 5 and 7"}],"minor_comments":[{"comment":"The sentence introducing Eq. (6) contains a typo: \"1 ER is the recoil energy\" should read \"where E_R is the recoil energy.\"","section":"Sec. III, Eq. (6)"},{"comment":"The title of Ref. [16] contains a typo: \"Freemen\" should be \"Freeman.\"","section":"Reference [16]"},{"comment":"The caption states that results are shown for the \"high-exposure (HE: σx = 15 nm, M = 100 g, right) scenario,\" but both panels of Fig. 12 are HE scenarios; the parenthetical \"right\" appears to be a leftover and should be removed or corrected.","section":"Fig. 12 caption"},{"comment":"The conditional nature of the mass reconstruction (the operator is assumed known) is stated only at the end of Sec. V; a one-sentence reminder in the caption of Fig. 3 would help prevent the contours from being read as unconditional mass measurements.","section":"Sec. V"},{"comment":"No code or data are provided, which limits reproducibility of the Asimov likelihood calculations; releasing the spectra and likelihood setup would strengthen the paper.","section":"General"}],"recommendation":"major_revision","confidential_remarks":"The paper leans heavily on the authors' own TFKS sensitivity projections both for setting signal normalizations and for bounding the parameter space shown in the contours. This is not circular in the statistical sense, but it means the reach claims are conditional on the accuracy of TFKS; I would encourage the editor to ensure that TFKS is available to the referees. I do not see any other concerns about novelty or scope."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear [Colleague],\n\nThis paper deserves a serious referee. It is the first detailed projection study of how well paleo-detectors can reconstruct WIMP mass and discriminate NREFT operators from track-length spectra, covering both elastic and inelastic scattering. The statistical framework is standard and applied transparently, and the authors clearly state the main assumptions: single-operator signals, fixed total event rate, operator-conditional mass reconstruction, and a deterministic energy-to-track-length mapping. Within those limits, the headline conclusions—mass reconstruction down to ~1–10 GeV with ~10% uncertainty, up to ~1 TeV at order-unity precision, and multi-sigma rejection of the standard SI/SD hypotheses for most non-canonical operators without directionality—are plausible and potentially valuable.\n\nThe main soft spot is the track-length mapping, Eq. (7), which every result inherits. The stress-test reader claims the authors admit this mapping breaks down in the low-mass regime where they claim their best reconstruction. That is not what the text says. Section VII explicitly states the one-to-one mapping is used for masses between 1 GeV/c^2 and 5 TeV/c^2 and is no longer valid only below 1 GeV/c^2, outside the claimed reach. So there is no internal contradiction. What is missing is a quantification of the residual spread in track length for fixed recoil energy and a demonstration that the mass contours and p-values are robust to it. That is a reasonable request, not a fatal flaw.\n\nTwo lesser points. Wilks' theorem is assumed for the profile likelihood ratio without Monte Carlo coverage checks; for Asimov data with 100 bins and mostly high occupancy it is likely fine, but a short coverage study would settle it. And no code or data accompany the paper, which makes independent verification harder, especially since much of the numerical input lives in the companion paper.\n\nThe paper is for paleo-detector practitioners and DM phenomenologists interested in near-future inference capabilities. It does not overclaim: it is a projection study, not a discovery, and the limitations are openly flagged. A serious editor should send this to peer review, with the reasonable request that the authors quantify the track-length mapping uncertainty and add minimal MC checks. The central argument holds up.\n\nYours","headline":"A careful, honest projection study whose central caveat—the deterministic track-length mapping—is real but not fatal, and was misread by the stress-test as an internal contradiction.","tokens_in":36368,"tokens_out":3719,"would_cite":true,"duration_ms":34606,"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":"Ancient mineral damage tracks could reconstruct WIMP masses from ~1 GeV to ~1 TeV and identify most non-standard dark-matter–nucleus interactions without directional read-out.","keywords":["dark matter direct detection","paleo-detectors","WIMP mass reconstruction","non-relativistic effective field theory","inelastic dark matter","nuclear recoil track lengths","profile likelihood ratio","mineral damage tracks"],"falsifier":"Irradiate polished gypsum and halite samples with monoenergetic ion beams spanning the recoil energies that WIMPs from 1 GeV/c² to 1 TeV/c² would produce, read out the resulting damage tracks at both 1 nm and 15 nm resolution, and compare the measured track-length distributions against Eq. (7); if the spread in track length at fixed recoil energy is comparable to or larger than the read-out resolution, the sharp spectral features that drive the projected ~10% mass uncertainties and the operator-discrimination significances would be smeared out, and the projections would not hold.","tokens_in":2072,"feed_emoji":"🪨","tokens_out":2407,"duration_ms":96417,"temperature":0.7,"pith_summary":"Paleo-detectors are ancient minerals whose accumulated crystal damage records nuclear recoils from dark-matter scattering over roughly a billion years. This paper asks what could be done with such a record if a WIMP signal were actually present, and answers that the track-length spectrum alone could reconstruct the WIMP mass for a broad set of interaction operators and could often identify which interaction produced the signal. It projects mass reconstruction down to about 10 GeV/c² with roughly 10% relative uncertainty in a high-resolution read-out, and up to about 1 TeV/c² in a high-exposure read-out, covering low-mass and inelastic regimes where conventional direct-detection experiments struggle. It further projects that the standard spin-independent or spin-dependent interaction could be rejected for nearly all non-standard operators at WIMP masses above about 10 GeV/c² without measuring recoil direction, which conventional experiments typically need.","feed_headline":"Ancient minerals could reveal dark matter masses down to 1 GeV","feed_subtitle":"Paleo-detectors project 10% mass precision at low WIMP masses and a 1 TeV reach, with no directional read-out needed.","key_machinery":"The working machinery is the binned track-length spectrum $dR/dx_T$, built by mapping each nuclear recoil energy to a track length through the stopping-power integral of Eq. (7), summing over the mineral's constituent nuclei, and convolving with a Gaussian read-out resolution. Signal spectra are computed for single isoscalar NREFT operators $O_i^s$, whose momentum- and velocity-dependent forms imprint different spectral shapes; the standard SI operator $O_1^s$ and SD operator $O_4^s$ serve as null hypotheses. Inference is done with a profile log-likelihood ratio on Poisson binned counts, using synthetic data sets and the asymptotic $\\chi^2$ distribution to define 2σ mass intervals and p-values for rejecting the standard-interaction hypothesis, with mineral mass, age, neutrino fluxes, and uranium concentration treated as constrained nuisance parameters. Two benchmark read-out scenarios—high resolution ($\\sigma_x = 1$ nm, 10 mg) and high exposure ($\\sigma_x = 15$ nm, 100 g)—span the resolution/exposure trade-off that drives which masses and operators can be distinguished.","core_discovery":"The central claim is that the shape of a paleo-detector's track-length spectrum—the histogram of damage lengths left by nuclear recoils—is sensitive enough to both WIMP mass and the operator structure of the WIMP–nucleus interaction to solve two inverse problems that conventional experiments find hard. For elastic scattering, a gypsum paleo-detector with 1 nm read-out resolution and 10 mg of sample is projected to reconstruct WIMP masses in the 1–10 GeV/c² range with relative uncertainty down to $\\Delta m_\\chi/M_\\chi \\sim 0.1$, and with 100 g of sample at 15 nm resolution to reconstruct masses up to ~1 TeV/c² with order-unity uncertainty for the standard SI operator and O$_{11}^s$. For inelastic scattering with a 50 keV/c² mass splitting, masses between 30 and 400 GeV/c² are reconstructible with relative uncertainties down to ~10⁻². The paper further claims that, given a signal generated by a non-standard NREFT operator, the standard SI hypothesis ($O_1^s$ in gypsum) or SD hypothesis ($O_4^s$ in halite) can be rejected at high confidence for nearly all such operators at WIMP masses above roughly 10 GeV/c², without any measurement of nuclear recoil direction; the exceptions are operators whose spectra are nearly identical to the standard ones, such as $O_8^s$ relative to $O_1^s$. The mass-reconstruction projections are conditional on the interaction operator being fixed or already identified; a fully simultaneous mass–coupling–operator fit is left to future work.","pith_inferences":["Because the paper normalizes every operator to the same total event rate, its discrimination claims isolate spectral shape; in a real detection the overall coupling must also be fitted, and that extra freedom will likely enlarge the projected exposure thresholds for rejection.","If the low-mass projections hold, the sub-10 GeV WIMP window could be probed with gram-scale mineral samples rather than ton-scale instruments, which would change the cost structure of low-mass direct detection.","The single-operator assumption probably makes both the mass intervals and the discrimination significances optimistic; realistic WIMP models mix operators with interference, and a joint fit could reveal degeneracies invisible in this analysis.","The one-to-one track-length mapping is directly testable with ion-beam calibrations; because the low-mass reach depends on short tracks near the resolution limit, those measurements would settle whether the reported ~10% precision is real."],"forward_implications":["In the high-resolution scenario ($\\sigma_x = 1$ nm, 10 mg), WIMP masses in the 1–10 GeV/c² range can be pinned down to $\\Delta m_\\chi/M_\\chi \\sim 0.1$, a range current conventional experiments cannot reach.","In the high-exposure scenario ($\\sigma_x = 15$ nm, 100 g), elastic mass reconstruction extends to 40–1000 GeV/c² with order-unity relative uncertainty for the standard SI operator and $O_{11}^s$, roughly doubling the mass range of analogous conventional-experiment studies.","For inelastic scattering with $\\delta m = 50$ keV/c², masses between 30 and 400 GeV/c² are reconstructible with relative uncertainties down to ~10⁻², a regime where conventional experiments lose sensitivity.","Standard SI ($O_1^s$) and SD ($O_4^s$) hypotheses can be rejected at more than 5σ for signals from $O_{15}^s$/$O_3^s$ in gypsum and $O_6^s$/$O_{13}^s$ in halite at masses above roughly 100 GeV/c² with exposures of order 1–100 kg·Myr, and no directional measurement is required.","Some operators remain degenerate: $O_8^s$ is nearly indistinguishable from $O_1^s$, and $O_7^s$ from $O_4^s$, so those specific non-standard interactions cannot be identified by track-length spectra alone."],"supporting_citations":[{"why":"The companion sensitivity study whose projected exclusion limits set the coupling normalizations and event rates used for every hypothetical signal.","marker":"[25]"},{"why":"The earlier paleo-detector mass-reconstruction analysis with the Euclideanized-signal method that this work generalizes to a profile-likelihood treatment over NREFT operators and inelastic scattering.","marker":"[19]"},{"why":"Supplies the accuracy check for the one-to-one track-length-to-recoil-energy mapping assumed in Eq. (7) over the 1 GeV/c² to 5 TeV/c² mass range.","marker":"[24]"},{"why":"Defines the non-relativistic effective operator basis and nuclear response functions from which all signal spectra are constructed.","marker":"[26, 27]"},{"why":"Provides the existing direct-detection exclusion limits that mark the parameter space already ruled out and set the conventional-experiment baseline.","marker":"[31]"},{"why":"Shows that conventional experiments need directional information to separate NREFT operators, the comparison point for the paper's direction-free discrimination claim.","marker":"[36]"},{"why":"Supplies the asymptotic profile-likelihood formalism used to define reconstructed mass intervals and rejection significances.","marker":"[79]"},{"why":"Established the track-length stopping-power description and the radiogenic-background model for paleo-detectors that the analysis builds on.","marker":"[18]"}],"fun_headline_variants":["Paleo-detectors reconstruct WIMP masses from 1 GeV to 1 TeV","Ancient minerals tell dark matter mass and operator","Paleo-detectors discriminate standard vs non-standard WIMP interactions","Paleo-detectors measure WIMP mass and operator without direction","Old crystals reconstruct dark matter mass and interaction without direction"],"cache_read_input_tokens":38400,"weakest_assumption_plain":"The projections rest on the assumption that a recoil of a given energy always produces the same track length, as computed from stopping powers; if real damage tracks have a spread of lengths at a fixed energy, the spectral features that carry both the mass measurement and the interaction identification would blur.","fun_headline_variants_meta":{"raw":{"variants":["Paleo-detectors reconstruct WIMP masses from 1 GeV to 1 TeV","Ancient minerals tell dark matter mass and operator","Paleo-detectors discriminate standard vs non-standard WIMP interactions","Paleo-detectors measure WIMP mass and operator without direction","Old crystals reconstruct dark matter mass and interaction without direction"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000443,"raw_usage":{"total_tokens":2390,"prompt_tokens":1241,"completion_tokens":1149,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":857,"completion_tokens_details":{"reasoning_tokens":1061}},"tokens_in":857,"tokens_out":1149,"duration_ms":10011,"temperature":1.0,"reasoning_tokens":1061,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T04:15:09.369935+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Irradiate polished gypsum and halite samples with monoenergetic ion beams spanning the recoil energies that WIMPs from 1 GeV/c² to 1 TeV/c² would produce, read out the resulting damage tracks at both 1 nm and 15 nm resolution, and compare the measured track-length distributions against Eq. (7); if the spread in track length at fixed recoil energy is comparable to or larger than the read-out resolution, the sharp spectral features that drive the projected ~10% mass uncertainties and the operator-discrimination significances would be smeared out, and the projections would not hold.","supporting_citations":[{"cited_title":"Bland-Hawthorn and O","cited_arxiv_id":null,"evidence_quote":"Supplies the asymptotic profile-likelihood formalism used to define reconstructed mass intervals and rejection significances."}],"review_version":1}