{"id":"8e0d2c01-35b2-47ae-8114-19fb8be6aa25","arxiv_id":"2504.21852","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"A hierarchical Bayesian framework, StarryStarryProcess, maps starspots from transit light curves and finds high-latitude spots plus significant spin-orbit misalignment on TOI-3884.","lead":"Astronomers built a probabilistic model that uses the brief dimming during exoplanet transits to map starspots on stellar surfaces, something rotation-only light curves cannot do well. Applied to TESS data for TOI-3884, the model reports high-latitude spots and a large spin-orbit misalignment, with implications for exoplanet characterization and stellar magnetic activity.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The claimed independent validation is undermined by three mutually inconsistent photometric v sin i values (Table 2: 1.69, §5.2: 5.2, Fig. 22 text: ~2.5 km/s) compared with the spectroscopic 3.59±0.92 km/s.","rationale":"The reader's weakest assumption is the small-perturbation linearization and the resulting inability to distinguish bright from dark features; that is a real limitation and the paper discloses it explicitly. In my reading, the more load-bearing problem is the internal inconsistency in the reported validation metric. The central claim—that transit photometry alone can recover 3D geometry and spot distributions—is supported by synthetic experiments, but the TOI-3884 application is the only external anchor. Section 5.2 explicitly uses v sin i agreement as independent validation, yet Table 2, §5.2, and Fig. 22 give three different values, and Table 2's rotation period disagrees with the literature. A reader cannot tell which posterior summary is correct, and the \"good agreement\" claim appears to select the favorable value. This does not necessarily invalidate the method, and the paper is unusually reproducible, which makes the proposed check feasible. The verdict should remain conditional: the method is promising, but acceptance of the TOI-3884 conclusions should require reconciling the v sin i and rotation-period inconsistencies and demonstrating that the polar-spot/obliquity result is robust to the literature rotation-period prior.","tokens_in":29228,"tokens_out":7159,"duration_ms":81402,"concrete_test":"Download the posterior chains from the Zenodo repository and recompute v sin i from the same posterior samples using the paper's assumed stellar radius and the relation v sin i = (2π R⋆/P⋆) sin i⋆, reproducing the summaries in Table 2, §5.2, and Fig. 22. Then rerun the TOI-3884 fit with a rotation-period prior consistent with Libby-Roberts et al. (e.g., P⋆ < 4.5 d) and check whether the near-polar spot and high-obliquity posteriors survive; if the three v sin i values remain mutually inconsistent or the polar spot shifts, the real-data conclusion is not supported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 5.2 states: \"Our derived stellar projected rotational velocity (v sini⋆ = 5.2+0.7−0.8 km/s) is in good agreement with spectroscopically measured values...\" and calls this \"strong independent validation of our geometric model.\" But Table 2 lists v sin i = 1.69+0.11−0.09 km/s, and the Figure 22 discussion says \"approximately 2.5 km/s\" and claims agreement with Libby-Roberts et al. (2023), who measure 3.59±0.92 km/s. These three numbers cannot all be the same posterior summary. This is not cosmetic: v sin i is the paper's principal external check on the photometrically recovered stellar inclination, obliquity, and rotation. The rotation period in Table 2 (P⋆ = 9.07+0.45−0.51 d) also conflicts with Libby-Roberts et al.'s <4.22±1.09 d. If the derived rotation/geometry is this fragile, the high-latitude spot concentration (µϕ ≈ 75°) and spin-orbit misalignment (ψ⋆ ≈ 80°) for TOI-3884 lose their independent support. The synthetic tests are encouraging, but they draw data from the same model and do not exercise the real-data degeneracies that make this validation necessary.","agreement_with_reader":"disagree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents StarryStarryProcess, a hierarchical Bayesian framework that combines the spherical-harmonic surface representation of starry with the Gaussian-process spot model of StarryProcess and a full transit model, in order to infer stellar surface features, stellar inclination, obliquity, and spot parameters from transit light curves. The surface map is marginalized analytically, leaving a Gaussian-process likelihood over the hyperparameters. The authors validate the framework with two synthetic experiments (a static surface and an evolving surface), recovering input parameters within roughly 2σ and illustrating the inclination–obliquity reflection degeneracy. They then apply the model to TESS observations of TOI-3884 and report a high-latitude spot concentration (µϕ ≈ 75°), a low stellar inclination (i⋆ ≈ 35°), and a large spin-orbit obliquity (ψ⋆ ≈ 80°). Section 5.2 claims that the photometrically derived v sin i provides strong independent validation of the geometric model. The paper has a reproducibility appendix, with code, data, and notebooks linked from the figures.","tokens_in":29603,"tokens_out":7805,"duration_ms":76758,"significance":"If the method is sound, the paper makes a useful methodological contribution: it extends the starry/StarryProcess framework to the transit-occultation geometry, shows that spot latitude, stellar inclination, and obliquity can in principle be constrained from single-band photometry, and demonstrates the approach on a real TESS target. The analytical marginalization of the surface map is elegant, the synthetic experiments are clearly described, and the reproducibility infrastructure (public code, chains, and notebooks) is a genuine strength. The central limitation is that the synthetic tests draw data from the same model family and therefore do not independently validate the TOI-3884 inferences; the claimed external validation through v sin i is, as written, internally inconsistent. The paper's significance for stellar-activity and obliquity studies depends on resolving that inconsistency and on clarifying the latitude-parameterization issue.","major_comments":[{"comment":"The manuscript reports three mutually incompatible values for the photometrically derived v sin i of TOI-3884: Table 2 gives 1.69+0.11−0.09 km/s, §5.2 states 'v sini⋆ = 5.2+0.7−0.8 km/s' and calls this 'strong independent validation', and the text around Figure 22 says 'approximately 2.5 km/s'. These cannot all be summaries of the same posterior. The comparison with the spectroscopic value of 3.59 ± 0.92 km/s from Libby-Roberts et al. (2023) is therefore not a validation; depending on which number is used, the photometric result sits on different sides of the spectroscopic measurement. Because v sin i is the principal external check on the photometrically recovered stellar inclination and obliquity, the TOI-3884 geometric claims need to be recomputed with a single, correctly defined conversion from P⋆ and i⋆, and the stellar radius used in that conversion must be stated consistently with §4.1.","section":"Table 2, §5.2, Figure 22"},{"comment":"The rotation period reported in Table 2, P⋆ = 9.07+0.45−0.51 d, is in strong tension with the Libby-Roberts et al. (2023) entry quoted in the same table, P⋆ < 4.22 ± 1.09 d, yet the discussion does not address this discrepancy. The rotation period and inclination jointly enter the v sin i derivation and the interpretation of spot latitudes, so a >2σ disagreement with an independent measurement is load-bearing and must be discussed, not omitted. Relatedly, §4.1 states that the stellar radius is fixed at 1 R☉; this is inconsistent with the inferred stellar density ρ⋆ = 15.18 g cm⁻³, which would correspond to a star of roughly 10.8 M☉ at 1 R☉. The radius assumption (or the sentence describing it) needs to be corrected, since the derived v sin i and the geometric interpretation depend directly on it.","section":"Table 2, §5.2, §4.1"},{"comment":"The latitude parameterization is defined through a Beta distribution on cosϕ, but Eqs. (6)–(7) identify µϕ and σ²ϕ with the mean and variance of a Beta random variable. For a random variable X = cosϕ, E[X] = α/(α+β) and Var[X] = αβ/((α+β)²(α+β+1)) are the mean and variance of cosϕ, not of the latitude ϕ. The text also refers to a 'mode' of the latitude distribution, which would have yet another formula. As written, the quoted TOI-3884 result 'µϕ = 75.24°' is therefore not the mean spot latitude under the stated model, and the interpretation of a near-polar spot is ambiguous until the correct mapping from (µϕ, σϕ) to (α, β) is supplied and used consistently in the figures and tables.","section":"Section 2, Eqs. (3)–(7)"},{"comment":"The synthetic experiments are self-consistency checks: the data are generated from the same StarryProcess prior and the same starry design matrix used for the inference, and the TOI-3884 analysis is the only independent test. Given that the v sin i validation in §5.2 is inconsistent, the paper currently lacks an out-of-sample check that would justify the claim that transit photometry alone can reliably recover stellar orientation and spot latitudes. The authors should either add a test with data drawn from a different generative model (for example, discrete circular spots, or a surface not obeying the GP prior) or explicitly restrict the validation claim to self-consistency.","section":"Section 3.2, Table 1"}],"minor_comments":[{"comment":"The prior on the planetary inclination is written as U(−bmax,−bmax); it should read U(−bmax, bmax).","section":"Section 3.1"},{"comment":"There are several typos that should be corrected in a revision: 'simultaniously' (§1), 'beacuse' (§2), 'distiguish' (§2), 'Firgure' (§3.3), 'imroved' (§5.3), and 'T able 1' in the Table 1 caption.","section":"Throughout"},{"comment":"The reproducibility software is cited as 'show your work! (?)'; the placeholder question mark should be replaced with a proper citation or a clear description of the tool.","section":"Appendix A"},{"comment":"The sentence 'yi+1 does not depend on yi−1' is confusing in the context of linear interpolation; the authors should clarify that the prior on each map is independent even though consecutive maps are interpolated between epochs.","section":"Section 2.3"}],"recommendation":"major_revision","confidential_remarks":"The core derivation is standard and the synthetic experiments are encouraging, but the TOI-3884 application has a load-bearing internal inconsistency: three different v sin i values are reported, and the rotation period tension with Libby-Roberts et al. (2023) is not addressed. The latitude-parameterization issue in Eqs. (3)–(7) also needs to be fixed before the polar-spot claim can be assessed. I do not see evidence of deliberate misreporting; the issues look like inconsistencies among posterior summaries computed with different conventions, but they are central to the paper's headline validation and must be resolved."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nHere's the short version: the core modeling idea is solid and worth engaging with, but the TOI-3884 validation section has internal inconsistencies that undermine the paper's central claim of independent verification.\n\nNew and good: StarryStarryProcess is a sensible hierarchical Bayesian integration of starry's spherical-harmonic light curve engine, StarryProcess's GP prior on spot distributions, and a full transit model that lets spot-crossing events inform the surface map. The synthetic experiments—both static and evolving—recover input parameters within ~2σ, and the paper makes code, data, and notebooks publicly available. That is real evidence that the machinery works in idealized conditions.\n\nSoft spots, in proportion: the big one is the photometric v sin i. Table 2 reports 1.69+0.11−0.09 km/s; Section 5.2 claims 5.2+0.7−0.8 km/s and calls it \"strong independent validation\"; the Figure 22 discussion says \"approximately 2.5 km/s.\" These cannot all be summaries of the same posterior. Since v sin i is the paper's principal external check on stellar inclination and obliquity, this inconsistency directly undercuts the validation narrative. The rotation period also conflicts with Libby-Roberts et al.'s <4.22±1.09 d vs. the reported 9.07+0.45−0.51 d, and the paper does not address that tension. The polar-spot conclusion (µϕ ≈ 75°) is more model-dependent than the text suggests: it leans on the Beta distribution prior on latitudes and on the small-spot linear approximation, and the hemisphere degeneracy is acknowledged but not fully resolved. These issues are fixable, but they need to be confronted before the TOI-3884 claims can be accepted.\n\nBottom line: this is a useful methodological contribution for transit spot mapping and JWST contamination work, and the code release is a plus. I'd send it to a serious referee, but with a clear request to reconcile the v sin i values and address the rotation-period discrepancy. If those are fixed, the paper will be a solid reference.","headline":"Solid new framework and synthetic validation, but the TOI-3884 validation claims are internally inconsistent—reconcile before accepting.","tokens_in":30084,"tokens_out":3986,"would_cite":true,"duration_ms":36861,"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":"Transit light curves alone can map starspot latitudes and recover stellar spin-orbit geometry.","keywords":["starspots","transit light curves","occultation mapping","spherical harmonics","Gaussian process","stellar obliquity","spin-orbit misalignment","TOI-3884"],"falsifier":"Generate synthetic transit light curves from surfaces with large, bright spots of order-unity contrast and run the same inference; if the recovered spot latitude distribution, size, or obliquity shifts outside the quoted uncertainties, the linearity assumption is falsified. Alternatively, multi-band transit photometry of TOI-3884 that resolves whether the crossing feature is bright or dark would directly test whether the near-polar spot interpretation is an artifact of the small-spot linear model.","tokens_in":29059,"feed_emoji":"🪐","tokens_out":7292,"duration_ms":70470,"temperature":0.7,"pith_summary":"This paper introduces StarryStarryProcess, a hierarchical Bayesian model that uses exoplanet transits as moving probes of a star's surface. It claims that by fitting the full photometric time series and modeling the small brightness bumps caused by spot crossings, single-band transit data can constrain the latitude distribution of starspots, the stellar inclination, and the obliquity between the stellar spin axis and the planetary orbit. The authors verify the recovery on synthetic data, extend the model to surfaces that evolve between transits, and apply it to TESS observations of TOI-3884, where they infer a near-polar spot concentration and a strongly misaligned orbit. If the claim holds, transit photometry becomes a practical tool for studying stellar magnetic activity and for separating stellar contamination from planetary signals in exoplanet observations.","feed_headline":"Transit spots expose TOI-3884's polar cap and tilted orbit","feed_subtitle":"Single-band TESS photometry recovers starspot latitudes, stellar inclination, and obliquity that usually require spectroscopy.","key_machinery":"The machinery is the Gaussian-process prior over spherical-harmonic surface coefficients, combined with the linear design matrix that turns a surface map into an observed light curve. The flux is written as $\\mathbf{M}(\\Theta)\\,\\mathbf{y}$, so with a Gaussian prior on the map $\\mathbf{y}$ the map can be integrated out exactly; the resulting marginal covariance $\\mathbf{B} = \\mathbf{C} + \\mathbf{M}\\Lambda\\mathbf{M}^\\top$ is what the sampler evaluates. Transits enter through the design matrix, so each spot-crossing bump contributes information along the planet's chord rather than only from disk-integrated rotation. A transformed stellar-orientation coordinate system with a half-normal prior breaks the reflection degeneracy between inclination and obliquity during sampling, and the time-dependent extension linearly interpolates between independent surface maps at successive epochs.","core_discovery":"The central discovery is that a Gaussian-process prior on the spherical-harmonic coefficients of a stellar surface, combined with an analytic transit design matrix, lets each transit carve information out of the null space that rotational light curves alone cannot see. Because both the prior and the map-to-flux operation are Gaussian and linear, the surface map can be marginalized analytically, leaving a Gaussian likelihood whose covariance encodes measurement noise plus spot-induced correlated variability. Spot-crossing events then constrain spot latitude, size, contrast, and number, along with stellar inclination and obliquity. For TOI-3884 the model finds spot latitudes concentrated near $\\pm 75^\\circ$, a stellar inclination of about $35^\\circ$, and a sky-projected obliquity near $80^\\circ$, with consistency against an independent spectroscopic $v \\sin i$ measurement.","pith_inferences":["If the method scales computationally, archival transit surveys could be mined for spot-latitude demographics, effectively turning exoplanet surveys into stellar-activity surveys.","Multi-band or spectroscopic transit observations, which make spot contrast wavelength-dependent, could break the paper's noted bright-versus-dark spot ambiguity and tighten the inferred spot sizes and latitudes.","The same analytic marginalization could be adapted to other linear mapping problems, such as eclipse mapping of binaries or Doppler tomography, wherever a Gaussian prior on the mapped quantity is reasonable.","The finding that spot-crossing statistics carry obliquity information suggests that future transit surveys could measure spin-orbit misalignment distributions for small planets, complementing traditional Rossiter-McLaughlin measurements."],"forward_implications":["Spot latitude distributions can be inferred from single-band transit photometry, not only from rotational modulation, so spot-crossing events become direct diagnostics of stellar magnetic activity.","Stellar inclination and obliquity can be constrained photometrically, opening spin-orbit studies for faint or otherwise inaccessible systems where high-resolution spectroscopy is impractical.","Jointly modeling spot crossings with transits reduces stellar contamination in derived planetary radii and transit shapes, supporting more accurate corrections for transmission-spectroscopy measurements.","The evolving-surface extension can track spot emergence, migration, and dissipation across multiple epochs, relevant to long-baseline surveys and future photometric missions.","For TOI-3884, the inferred polar spot concentration and near-$80^\\circ$ obliquity constitute a photometric signature of spin-orbit misalignment and high-latitude magnetic flux emergence."],"supporting_citations":[{"why":"Supplies the spherical-harmonic design matrix used to compute transit and rotational light curves from surface maps.","marker":"Luger et al. 2019"},{"why":"Supplies the Gaussian-process prior on spherical-harmonic coefficients, the statistical spot model this paper builds on.","marker":"Luger et al. 2021a"},{"why":"Introduced the idea that a transiting planet scans the stellar surface and can be used to map spots.","marker":"Silva 2003"},{"why":"Provides the marginalization identities and efficient inverse that let the model integrate out the surface map.","marker":"Hogg et al. 2020"},{"why":"Provides independent TOI-3884 radial-velocity, stellar, and orbital constraints used for priors and comparison.","marker":"Libby-Roberts et al. 2023"},{"why":"Discovered TOI-3884b and supplied initial transit parameters adopted as priors.","marker":"Almenara et al. 2022"},{"why":"Supplies the parallel-tempered MCMC sampler used to handle the multimodal posterior.","marker":"Vousden et al. 2016"},{"why":"Supplies the data-processing tools used to extract and clean the TESS light curves.","marker":"Lightkurve Collaboration et al. 2018"},{"why":"Provides the magnetohydrodynamic expectation of polar spot emergence cited to interpret the high-latitude result.","marker":"Yadav et al. 2015"}],"fun_headline_variants":["Transit mapping reveals TOI-3884's polar spots and tilt","How transits expose hidden starspot latitudes on TOI-3884","TOI-3884's spots and orbit tilt decoded via transits","Transit light curves map stellar surfaces: TOI-3884 case","New model uses transits to find polar spots on TOI-3884"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The model assumes spot-induced brightness changes are small enough that the flux is linear in the spherical-harmonic surface map, an assumption the paper notes prevents it from distinguishing bright from dark spots; if spots are large or include bright regions, the inferred contrasts, sizes, and the high-latitude reading of TOI-3884 could be biased.","fun_headline_variants_meta":{"raw":{"variants":["Transit mapping reveals TOI-3884's polar spots and tilt","How transits expose hidden starspot latitudes on TOI-3884","TOI-3884's spots and orbit tilt decoded via transits","Transit light curves map stellar surfaces: TOI-3884 case","New model uses transits to find polar spots on TOI-3884"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000223,"raw_usage":{"total_tokens":1472,"prompt_tokens":975,"completion_tokens":497,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":591,"completion_tokens_details":{"reasoning_tokens":400}},"tokens_in":591,"tokens_out":497,"duration_ms":4763,"temperature":1.0,"reasoning_tokens":400,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T04:52:49.973143+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Generate synthetic transit light curves from surfaces with large, bright spots of order-unity contrast and run the same inference; if the recovered spot latitude distribution, size, or obliquity shifts outside the quoted uncertainties, the linearity assumption is falsified. Alternatively, multi-band transit photometry of TOI-3884 that resolves whether the crossing feature is bright or dark would directly test whether the near-polar spot interpretation is an artifact of the small-spot linear model.","supporting_citations":[],"review_version":1}