{"id":"74f1640b-a0da-4b25-bcc0-0a1f7124ff5e","arxiv_id":"2607.20599","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":8,"one_line_summary":"A wind-reprocessing model of unstable mass transfer onto neutron stars and black holes is fit to the long-rising precursors of SN 2023zkd, SN 2023fyq, and SN 2021qqp, inferring mass-loss histories consistent with binary merger progenitors.","lead":"This paper builds a semi-analytical model for the slow-brightening light emitted by mass transfer onto a compact star before a binary merger, and fits it to three supernova precursor events. The model lets astronomers turn long-rising precursor light curves into constraints on the mass-transfer history and the donor star.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Inferred mass-transfer histories are conditioned on the assumed two-stage form of Mdot(t) (Eq. 6); without synthetic-data validation against alternative Mdot(t) shapes, the capability claim is not yet demonstrated.","rationale":"The reader's weakest_assumption correctly identifies the assumed functional form of Mdot(t) and the spherical wind geometry as the principal vulnerability. My stress-test converges on the same point: the inference of mass-transfer history is only as good as the parametric family in Eq. (6), and the paper does not demonstrate through synthetic recovery tests that this family is expressive enough to capture plausible alternative histories. The paper is honest about the approximation (Section 2.1) and about geometrical uncertainties (Section 3.3), and the authors present a public code and three applications—genuine contributions. However, the abstract's phrase 'demonstrating the model's capability of inferring the mass-transfer history' overstates what has been shown: fitting three events with a flexible 8-parameter model does not by itself establish that the inferred histories are robust to the assumed shape. A synthetic-data injection test would settle this directly. Because this is a validation gap rather than an internal inconsistency, the reader's CONDITIONAL verdict remains appropriate; I would not move to REJECT or UNVERDICTED. The condition should be explicit: demonstrate recovery of a known Mdot(t) outside the Eq. (6) family, or clearly restate the claim as 'fitting within a prescribed mass-transfer model' rather than 'inferring the history.'","tokens_in":26920,"tokens_out":4626,"duration_ms":42612,"concrete_test":"Generate synthetic precursor light curves and temperatures from a known Mdot(t) that deliberately does not follow Eq. (6)—e.g., Mdot(t) = Mdot0*(1 + exp((t−t0)/τ)) or a broken power law with δ changing at some epoch—matching the cadence and noise of the SN 2023zkd data. Feed these into the public MCMC code using the Eq. (6) model. If the recovered median Mdot(t) deviates from the true history by more than the 1σ posterior envelope, or if the posterior simply locks onto the closest Eq. (6) shape, then the inference of mass-transfer history is shape-dominated and the central claim must be qualified.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim—that the model can 'infer the mass-transfer history' (abstract; §5)—rests on the parametric form of Mdot(t) adopted in Eq. (6), which fixes the history as a constant plateau followed by a single power-law divergence ∝(t_m−t)^−δ. The MCMC inference explores only this five-parameter family, so the 'reconstructed' Mdot(t) is not a free-form inference but a projection onto a prescribed curve. If the true mass-transfer history has a different early-time dependence—e.g., a thermal-timescale exponential rise, a secondary plateau, or a different divergence exponent—the posteriors on (Mdot0, t0, δ, tm) will shift to compensate, biasing the donor properties derived through Eq. (7) in §4.4. The paper itself concedes the approximate nature of Eq. (1) in §2.1 (valid only when the Roche-filling fraction is small) and notes in §3.3 that L2 outflow geometry can be equatorial, which would break the quasi-spherical wind assumption underlying the L-T mapping (§2.3–2.4). §5 further states that the work is 'limited to an inverse problem approach' with forward modeling deferred. No synthetic-data validation is presented to show that the assumed shape is recoverable or that the posteriors are not dominated by the functional prior. This is the most load-bearing soft spot because the strongest claim is about inference capability, not just about producing a model that can fit three events.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This manuscript constructs a semi-analytical model for long-rising optical precursors of binary mergers with compact-object accretors. The model assumes a quasi-spherical, optically thick wind reprocessing accretion power; the mass-transfer history is parameterized as a constant plateau followed by a power-law divergence (Eq. 6). It includes dust extinction and computes luminosity and temperature, with eight free parameters. The model is fit via MCMC to three observed SN precursors (SN 2023zkd, 2023fyq, 2021qqp), yielding reconstructed mass-transfer histories and progenitor interpretations. The code is publicly available.","tokens_in":27396,"tokens_out":7691,"duration_ms":63005,"significance":"The paper addresses a timely problem: interpreting a growing sample of long-rising precursors of interacting SNe as unstable mass transfer onto compact objects. Its strengths include a transparent, semi-analytical formalism; a public implementation; joint use of luminosity and temperature to break the Mdot-epsilon degeneracy; and honest confrontation with external data (e.g., CSM mass for 2023zkd; radio mass-loss estimates for 2023fyq). If the inference is robust, this is a valuable framework for LSST-era events. The principal caveat is that the 'inferred mass-transfer history' is a projection onto the assumed functional form of Eq. (6), and no synthetic-data test demonstrates recovery of other shapes. This is addressable and does not invalidate the model as a forward model.","major_comments":[{"comment":"The central claim of the abstract and §5 — that the method infers the mass-transfer history — is conditioned on the assumed two-stage form of |Mdot(t)| in Eq. (6). The MCMC only explores the parameters of this prescribed curve, so the reconstructed Mdot(t) is not a free-form inference. This matters because §4.4 uses the reconstructed Mdot in Eq. (7) to derive donor properties (e.g., R*~100 Rsun, M*~10s Msun for the Type IIn events; He-star+NS for 2023fyq). No synthetic-data experiment is shown in which an Mdot(t) with a different early-time behavior is injected and recovered. I recommend either adding such a validation (e.g., injecting a two-plateau or exponential-rise history and checking posteriors) or explicitly limiting the claim in the abstract and §5 to 'constraining parameters within the assumed family.' This is the most load-bearing point.","section":"§2.1, Eq. (6); §5"},{"comment":"For SN 2023zkd, the fit uses five blackbody epochs (Figure 5; ten scalar measurements) against eight free parameters (Table 2). The corner plot (Figure 10) shows weak constraints on tMT (-19.3 +7.6/-5.8 yr) and broad posteriors on several parameters. The reconstructed history in Figure 6 is therefore strongly influenced by the prior shape of Eq. (6). A cross-validation check, an information criterion, or a statement of effective degrees of freedom would make the 'capability' claim more convincing. This is related to Major Comment 1 and could be addressed in the same validation.","section":"§4.1, Table 2, Figure 5"},{"comment":"The luminosity-temperature mapping assumes a quasi-spherical wind, but §3.3 acknowledges that the L2 outflow may be equatorial and that the light curve could then be two-component. The current version does not quantify how much an asymmetric geometry biases the inferred Mdot(t) or the donor properties in §4.4. Because the abstract claims a general capability, the authors should either add a simple anisotropy study or qualify the claim in the abstract and §5 to the quasi-spherical case.","section":"§2.3–2.4, §3.3"}],"minor_comments":[{"comment":"Duplicate 'our' in 'our our light curve model'.","section":"§4"},{"comment":"Typo: 'currect model' should be 'current model'.","section":"§5"},{"comment":"The prior range for tMT appears as '[-104 days, first detection]'; should read '[-10^4 days, first detection]'.","section":"Table 2"},{"comment":"Kramer's opacity' should be 'Kramers' opacity' for consistency with §2.6.","section":"Eq. (25)"},{"comment":"The g-band residuals near t≈-1.8 yr are attributed to weighting toward the many ramp-up points; showing the weights or residuals explicitly would improve transparency.","section":"Figure 9"},{"comment":"The inverse formula uses the positive branch of the square root; the authors should state the range of color over which this branch is valid.","section":"Eq. (33)"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is within the journal's scope. The main gap is the mismatch between the strong 'inference capability' claim and the parametric prior in Eq. (6); I think a synthetic-recovery test plus language softening will be sufficient. No concerns about novelty or citation practice."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a solid contribution that deserves a serious referee, with one soft spot that should be fixed before publication. The model and code give the community a workable tool for LSST-era precursors, and the three applications show genuine care for the data. But the paper's headline capability—inferring the mass-transfer history—is currently only demonstrated within a prescribed two-parameter family (plateau plus power-law rise). Without synthetic-data tests against other shapes, that claim is more limited than the abstract suggests.\n\nThe new piece relative to Piro & Lu (2020) and Tsuna et al. (2024) is the time-dependent mass-transfer prescription, dust treatment, and the public MCMC inference framework. The temperature information really does help break the Mdot-epsilon degeneracy, as shown clearly in Figure 6. The authors are also honest with their external checks: the CSM mass for 2023zkd comes out comparable to the SN-phase estimate without being imposed, and the radio mass-loss tension for 2023fyq is discussed fairly rather than hidden.\n\nThe main issue is the parametric form of Mdot(t) in Eq. (6). The MCMC explores only that family, so the 'reconstructed' history is a projection onto a prescribed curve. The authors acknowledge in Section 5 that this is an inverse-problem approach with forward modeling deferred, and they note the geometry and validity limitations in Sections 2.1 and 3.3. What is missing is a synthetic-data validation: generate light curves from a different Mdot(t) (e.g., exponential rise or a second plateau) and show whether the posteriors recover the true integrated mass or bias it. That test would tell us whether the framework constrains physics or just the chosen parameters. For 2023zkd, five blackbody epochs for eight parameters is thin; the posteriors are broad enough that this is more of a demonstration than a measurement. The velocity prior for 2023fyq is a bit circular, since the same wind model produces the velocity being constrained, but the prior is placed at a specific epoch and the posterior is much tighter than the prior, so this is minor.\n\nOverall the derivation is internally consistent, the code and data are public, and the limitations are stated rather than buried. I would not hold the soft spots against the central value of the paper, which is to give observers a concrete framework for interpreting these precursors.\n\nRecommendation: send to peer review. Require synthetic-data recovery tests and a more explicit statement that inferred histories are conditional on the assumed functional form. Then publish. I'd cite this when working on precursor transients.","headline":"A useful and honest semi-analytical framework for SN precursors, but the 'mass-transfer history inference' is really a fit to an assumed two-stage shape until synthetic-data validation shows otherwise.","tokens_in":27781,"tokens_out":2299,"would_cite":true,"duration_ms":20773,"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":"A semi-analytical model reconstructs the mass-transfer history of merging binaries from years-long supernova precursor light curves and temperatures.","keywords":["supernova precursors","binary mergers","unstable mass transfer","accretion power","compact object companions","light curve modeling","wind reprocessing","transient surveys"],"falsifier":"The model predicts that a precursor's color temperature should fall as the wind becomes denser (constant or rising luminosity with falling temperature). A precursor whose multi-band photometry shows temperature rising alongside luminosity, or whose radio-inferred mass-loss rate differs from the model-reconstructed Mdot by much more than the factor of a few the paper already attributes to asymmetry in SN 2023fyq, would contradict the quasi-spherical, optically thick wind assumption.","tokens_in":26838,"feed_emoji":"💫","tokens_out":9617,"duration_ms":88200,"temperature":0.7,"pith_summary":"The paper argues that the long-rising precursor emission seen before some interacting supernovae is accretion-powered light reprocessed by a quasi-spherical, optically thick wind launched by a compact object companion during unstable mass transfer. It constructs a semi-analytical model that turns the observed luminosity and temperature into a time-resolved mass-transfer rate, assuming the rate stays constant for a while and then rises as a power law toward merger. Applied to three events, the model recovers mass-transfer rates of order 0.1–1 solar masses per year in the final years, favors evolved giant donors with black hole or neutron star companions, and shows that temperature data are what break the degeneracy between mass-loss rate and accretion efficiency. If correct, this gives a direct way to watch the last years of a merging binary and to identify a population of merger-driven explosions in upcoming wide-field surveys.","feed_headline":"Precursor light curves decode binary mergers years before the blast","feed_subtitle":"Fitting luminosity and temperature recovers each donor's mass-loss history years before merger","key_machinery":"The load-bearing object is the quasi-spherical optically thick wind. Two radii set the emission: the trapping radius, where the diffusion time equals expansion time and most luminosity escapes, and the thermalization radius, where radiation and gas decouple (found with a bound-free/free-free absorption opacity with a recombination cutoff); the observed temperature is the wind temperature at the larger radius. The mass-transfer history is prescribed as a constant rate followed by a power-law rise (t_m - t)^(-δ), with δ tied to the donor's envelope structure, and dust reddening is added through a sublimation radius. This maps observed luminosity and temperature to the binary's mass-transfer ra","core_discovery":"The paper's central claim is that years-long supernova precursors are powered by unstable mass transfer onto a compact object, and that the light curve is the emission of an optically thick wind reprocessing accretion power. Solving the wind shell by shell shows luminosity is set at the trapping radius; temperature is set at the thermalization radius. Because temperature traces mass-transfer rate divided by wind velocity while luminosity traces their product, the pair breaks the degeneracy between them. Applied to three real precursors, the model recovers mass-transfer histories that rise toward merger with total masses matching the inferred circumstellar material.","pith_inferences":["Editorial: if the assumed two-stage mass-transfer law is generic, the shape of the late rising light curve directly measures the donor envelope's polytropic index (δ = 1 + 2/(2n+1)), turning a survey of precursors into a population-level measurement of donor structure.","Editorial: the model's dust treatment implies a sharp drop in optical dust optical depth as the precursor brightens near merger—an infrared signature that could be tested with targeted follow-up observations of future precursors.","Editorial: the same wind-reprocessing framework could be adapted to other long-rising transients (e.g., luminous red novae), but for those the donor's own emission and gas pressure would need to be included, as the authors themselves note."],"forward_implications":["Precursor light curves of interacting supernovae can be inverted to recover the mass-transfer history of the pre-merger binary, giving the mass-loss rate and total mass shed in the final years.","Temperature information (multi-band photometry or spectroscopy) is what breaks the degeneracy between mass-transfer rate and accretion efficiency; bolometric-only fits leave the mass-loss history poorly constrained.","The three fitted events favor evolved, giant-like donors with black hole companions for the Type IIn cases and a low-mass helium star with a neutron star for the Type Ibn case.","The double-peaked morphology shared by these events is explained as a delay between the compact object's plunge-in and the final explosion, with a prediction of at most two well-separated peaks.","The model applies directly to the large sample of precursors expected from next-generation sky surveys, where dimmer events with lower-mass donors or neutron star companions should be far more numerous."],"fun_headline_variants":["Light curve model decodes pre-merger accretion","Pre-merger glow reveals mass transfer history","Accretion-powered precursors map binary fate","Years-long precursors hint at merger explosion","Wind-driven light curves trace binary death spiral"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The model assumes the erupted material forms a roughly spherical, opaque wind whose flow rate stays flat for years and then surges as a single steepening curve right up to the moment of merger; if the flow is lumpy, lopsided, or ramps up differently early on, every inferred mass-transfer history and donor property would be off.","fun_headline_variants_meta":{"raw":{"variants":["Light curve model decodes pre-merger accretion","Pre-merger glow reveals mass transfer history","Accretion-powered precursors map binary fate","Years-long precursors hint at merger explosion","Wind-driven light curves trace binary death spiral"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000167,"raw_usage":{"total_tokens":1056,"prompt_tokens":667,"completion_tokens":389,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":411,"completion_tokens_details":{"reasoning_tokens":337}},"tokens_in":411,"tokens_out":389,"duration_ms":4836,"temperature":1.0,"reasoning_tokens":337,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-01T09:49:13.335286+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"The model predicts that a precursor's color temperature should fall as the wind becomes denser (constant or rising luminosity with falling temperature). A precursor whose multi-band photometry shows temperature rising alongside luminosity, or whose radio-inferred mass-loss rate differs from the model-reconstructed Mdot by much more than the factor of a few the paper already attributes to asymmetry in SN 2023fyq, would contradict the quasi-spherical, optically thick wind assumption.","supporting_citations":[],"review_version":1}