{"id":"3f6e2ba3-b443-4e91-a72c-4fbeae276736","arxiv_id":"2608.13003","paper_version":1,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"low","formal_verification":"none","parameter_count":1,"one_line_summary":"A new luminous fast blue optical transient, AT 2024qfm, was discovered at z=0.2267 with a Lasair broker filter and matches AT 2018cow in luminosity and 13 day fade.","lead":"Astronomers report the discovery of AT 2024qfm, a fast, bright, blue cosmic flash classified as a luminous fast blue optical transient (LFBOT) at redshift 0.2267. It is nearly identical in brightness and fade speed to the prototype AT 2018cow and was found with a new broker filter that flags rapidly fading sources.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"K-correction sign error inflates AT 2024qfm's peak absolute magnitude: reported M_g=-21.0 should be about -20.5, affecting the central luminosity claim.","rationale":"The reader's weakest assumption (host-galaxy redshift) is reasonable but well supported: the Gemini/GMOS spectrum shows narrow nebular emission lines at a common redshift that extend to the transient position, with a velocity offset (+99 km/s) consistent with the host's rotation. A chance superposition would require a remarkable coincidence, so this is not the most insecure link. A more concrete and demonstrable issue is the K-correction sign error. The paper's own data and stated cosmology yield Mg=-20.6, not -21.0, if the standard relation m = M + DM + K is used. The reported -21.0 is exactly what one obtains by adding the negative K to m-DM instead of subtracting it. This is an internal inconsistency, checkable from Table A1 alone, and it directly affects the paper's strongest quantitative claim and its comparison to AT 2018cow. The error does not invalidate the LFBOT classification, since -20.6 still lies above the class threshold, but it does require revision of the absolute magnitudes and the 'almost identical' statement. Therefore the appropriate verdict is conditional acceptance pending correction of the K-correction sign and re-evaluation of Figure 4 and the derived luminosities.","tokens_in":14804,"tokens_out":20623,"duration_ms":187137,"concrete_test":"Recompute AT 2024qfm's absolute magnitude from Table A1 using the standard K-correction convention: M = m - DM - K, with K = -2.5log10(1+z) and DM = 5log10(d_L/10pc) for the stated cosmology (H0=70, Omega_m=0.3). At z=0.2267 this gives DM≈40.26 and K=-0.245; applying to the peak r-band magnitude m=19.41 yields M≈-20.6 rather than -21.0. If the recomputed value differs from the paper's -21.0 by ~0.5 mag, the sign error is confirmed. Also recompute the comparison points for AT 2020xnd and AT 2023fhn using the same correction to assess the similarity claim.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"Section 5 applies 'a correction of K=-2.5 log(1+z) is added when calculating the absolute magnitude' for AT 2024qfm (and AT 2020xnd, AT 2023fhn). Under the standard convention (Hogg et al. 2002), m = M + DM + K, so M = m - DM - K. With K = -2.5log(1+z), the correct expression is M = m - DM + 2.5log(1+z). If instead the negative K is added to m-DM, the result is M = m - DM - 2.5log(1+z), which is 5log(1+z) ≈ 0.49 mag too bright at z=0.2267. Using the paper's own Table A1: peak r-band magnitude m_r=19.41 and distance modulus DM≈40.26, the correct absolute magnitude is M_g = 19.41 - 40.26 + 0.245 = -20.6, not -21.0. The paper's value of Mg=-21.0 follows from the incorrect sign (19.41 - 40.26 - 0.245 = -21.1). The same error affects AT 2020xnd and AT 2023fhn, and because AT 2018cow and AT 2024wpp were plotted without K-corrections, the comparison in Figure 4 is inconsistent: AT 2024qfm is artificially brightened by ~0.5 mag relative to AT 2018cow, strengthening the claim of similarity. The classification as an LFBOT likely survives (Mg=-20.6 still exceeds the -20 threshold), but the headline quantitative claim and comparison require correction.","agreement_with_reader":"disagree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper reports the discovery and follow-up of AT 2024qfm, a rapidly evolving blue transient at z = 0.2267 that the authors classify as a luminous fast blue optical transient (LFBOT). The transient was identified in the ZTF alert stream by the Lasair FastFinder annotator on the basis of a fast decline and blue colour. The authors present multi-band photometry from ZTF, ATLAS, Pan-STARRS, LOT, LT, and Swift UVOT, together with Gemini/GMOS and NOT/ALFOSC spectroscopy, and use the host-galaxy redshift from narrow nebular lines to derive a peak absolute magnitude M_g ≈ -21.0. They compare AT 2024qfm with AT 2018cow and other LFBOTs, fit the host SED with Bagpipes, and discuss how the FastFinder approach could be applied to LSST data. The central claim is that AT 2024qfm is a member of the small LFBOT sample, closely resembling AT 2018cow in luminosity, colour, and spectral appearance.","tokens_in":15172,"tokens_out":9982,"duration_ms":94403,"significance":"If the quantitative results are corrected as detailed below, this paper is a valuable contribution: it adds a well-observed member to the small LFBOT sample, demonstrates a practical broker-based selection method for fast transients, and releases photometric tables and public spectra. The classification as an LFBOT is based on direct photometric and spectroscopic evidence rather than on model fitting, and the fast-rise/fast-decline plus blue-colour selection is a legitimate discovery strategy rather than circular reasoning. The host-galaxy redshift is secured by multiple narrow emission lines that extend to the transient position, making the physical association credible. The main quantitative claim, M_g ≈ -21.0, requires revision because the K-correction is applied with the wrong sign; the corrected value is about -20.6, which still places the object above the nominal LFBOT luminosity threshold but weakens the quantitative similarity to AT 2018cow.","major_comments":[{"comment":"The K-correction is applied with the wrong sign. The paper states that 'a correction of K = -2.5 log(1+z) is added when calculating the absolute magnitude'. Under the Hogg et al. (2002) convention used in the text, m = M + DM + K, so M = m - DM - K. With K = -2.5 log(1+z), the correct expression is M = m - DM + 2.5 log(1+z). If the negative K is instead added to m - DM, the result is M = m - DM - 2.5 log(1+z), which makes the source 5 log(1+z) ≈ 0.44 mag too bright at z = 0.2267. Using the Table A1 peak r-band magnitude m_r = 19.41 and the adopted cosmology (DM ≈ 40.26), the correct rest-frame g-band absolute magnitude is 19.41 - 40.26 + 0.222 = -20.6, not -21.0. The same sign error affects the quoted absolute magnitudes of AT 2020xnd and AT 2023fhn. Please correct the sign, recompute the affected absolute magnitudes and Figure 4, and update the statements in Sections 5 and 6 that report M_g = -21.0.","section":"Section 5, Section 6, Table A1"},{"comment":"The comparison sample in Figure 4 is computed in inconsistent magnitude systems. AT 2018cow and AT 2024wpp are plotted without any colour or K-correction, while AT 2024qfm, AT 2020xnd, and AT 2023fhn receive the approximate K-correction. With the sign error above, AT 2024qfm appears about 0.44 mag brighter than it should be, while AT 2018cow (z ≈ 0.014) is essentially unshifted. The conclusion that the rest-frame ugri light curves are 'remarkably similar' and that AT 2024qfm is 'almost identical' to AT 2018cow is therefore partly an artefact of the heterogeneous correction scheme. Please recompute all objects in a single consistent rest-frame system, or apply the same approximate K-correction to every object, and reassess the similarity claim.","section":"Figure 4 and Section 5"}],"minor_comments":[{"comment":"The lack of template subtraction for the Swift UVOT photometry is acknowledged, but the possible host-galaxy contamination is not quantified. Since the UV points are used in the comparison with AT 2018cow, please add an estimate of the host contribution at the UVOT epochs or explicitly justify that it is negligible.","section":"Section 3.1"},{"comment":"Table A1 states that all measurements are 'uncorrected for Galactic or host dust extinction', while the Figure 2 caption says that magnitudes have been corrected for Galactic extinction. Please clarify which values are plotted and which are used for the absolute-magnitude calculations.","section":"Appendix A and Figure 2"},{"comment":"The decline-rate sign convention is inconsistent: dg/dt = +0.37 mag/d for AT 2024qfm but dmg/dt = -0.26 mag/d for AT 2024kth. Please use a single convention throughout.","section":"Section 6"},{"comment":"The host-galaxy association is critical for all luminosity estimates. The evidence presented is reasonable, but a quantitative chance-coincidence estimate or an explicit statement of the spatial and velocity coincidence criteria would strengthen the paper.","section":"Section 3.2"},{"comment":"There are typographical errors: 'analyse the the host galaxy' in Section 4 and 'lower than than that' in Section 5. Please correct them.","section":"Sections 4 and 5"}],"recommendation":"major_revision","confidential_remarks":"The K-correction sign error is the main issue; it is local and fixable, and the LFBOT classification appears to survive the correction. No concerns about the novelty or scope of the paper for MNRAS. I would recommend acceptance after the revision."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Fulton et al. report AT 2024qfm, a new LFBOT at z=0.2267, found through the FastFinder annotator on Lasair-ZTF. That is the real news here: a fast-transient discovery method that could plausibly scale to LSST, demonstrated on a live alert stream. The paper is also refreshingly honest, stating up front that the UV/optical data do not yield new physical insight into LFBOTs. The classification itself is on solid ground: secure host redshift from multiple narrow nebular lines, a fast rise and decline, blue featureless spectra, and an absolute magnitude that remains luminous even after my correction below.\n\nThe photometry is carefully reduced, ground-based data are template subtracted, and the Swift UVOT contamination is acknowledged rather than hidden. The data availability statement is strong, and the FastFinder description, while tied to a thesis, is concrete enough to be useful. This deserves to be published as a sample-building paper.\n\nThe soft spot is quantitative. Section 5 says a K-correction of K=-2.5log(1+z) is 'added when calculating the absolute magnitude.' With the standard Hogg et al. convention, M = m - DM - K, so adding a negative K brightens the source by 5log(1+z). For z=0.2267 that is about 0.49 mag. Using the paper's own peak r-band magnitude and DM, the correct rest-frame g-band absolute magnitude is roughly -20.6, not -21.0. The same sign error affects AT 2020xnd and AT 2023fhn in the comparison, and since AT 2018cow and AT 2024wpp were plotted without K-corrections, Figure 4 mixes conventions. The LFBOT classification survives (M_g=-20.6 still passes the usual -20 threshold), but the headline luminosity claim and the quantitative comparison in Figure 4 are off by half a magnitude. That is a load-bearing number for at least part of the paper's value, so it needs to be fixed, not just footnoted.\n\nThe LSST forecasting also assumes M_g=-21; shifting to -20.6 will alter the stated volume limits modestly, which is worth a check.\n\nThis is exactly the kind of discovery paper that belongs in the literature after revision. The qualitative conclusions are sound, the method is interesting, and the error is correctable. I would send it to a referee, with the clear instruction to verify the K-correction sign and the resulting absolute magnitudes.","headline":"A solid, useful LFBOT discovery paper with a real K-correction sign error that inflates the peak absolute magnitude by half a magnitude and needs fixing before publication.","tokens_in":753,"tokens_out":885,"would_cite":true,"duration_ms":38602,"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":"AT 2024qfm is a luminous fast blue optical transient at z=0.2267 that closely matches AT 2018cow, and a flux-gradient alert filter found it in the ZTF alert stream.","keywords":["LFBOT","fast blue optical transient","AT 2024qfm","Zwicky Transient Facility","Lasair broker","alert-stream filtering","AT 2018cow","LSST"],"falsifier":"Take a spatially resolved spectrum of the host across the transient position: if the narrow nebular lines at the transient are not at the same systemic velocity as the host, or come from a background galaxy, the redshift association—and with it the absolute magnitude and LFBOT classification—breaks.","tokens_in":14645,"feed_emoji":"⚡","tokens_out":9140,"duration_ms":79927,"temperature":0.7,"pith_summary":"AT 2024qfm is a new member of the small, rare class of luminous fast blue optical transients (LFBOTs), objects that brighten and fade within days, stay blue and featureless in the optical, and outshine normal supernovae at peak. The paper makes the case that this transient, at redshift z=0.2267, is observationally almost identical to the class prototype AT 2018cow: similar peak absolute magnitude (M_g≈-21), similar rapid ~13-day decline, and a blue featureless spectrum near peak. It also demonstrates a practical path to finding more such events: a custom annotator running on the Lasair broker flags candidates by their large flux gradients and blue colour in the Zwicky Transient Facility alert stream. With the coming Rubin Observatory LSST, the same approach could expand the search volume tenfold and yield roughly one LFBOT per month, provided the fast evolution can be caught early enough for follow-up.","feed_headline":"Fast blue optical transient AT 2024qfm is a twin of AT 2018cow","feed_subtitle":"The ZTF/Lasair FastFinder filter caught a 13-day, blue, M_g=-21 transient; LSST could find one per month.","key_machinery":"The central object is the FastFinder annotator running on the Lasair broker: real-time code that scores ZTF alerts by the gradient of their brightness over time (both rise and decline) and by colour, flagging a fast decline (dg/dt = 0.37±0.14 mag $d^{-1}$) and blue colour as the signatures of an LFBOT. This, together with the observational signature of LFBOTs defined by AT 2018cow—rapid rise and decline on few-day timescales, blue featureless continuum, and peak M_g≈-21—carries the argument: the match of AT 2024qfm's multi-band lightcurve and spectra to AT 2018cow places it unambiguously in the class. The host-galaxy redshift from narrow nebular lines provides the distance scale that converts apparent to absolute magnitude.","core_discovery":"The paper establishes that AT 2024qfm is an LFBOT: its rise and decline are confined to a few days, its spectrum near peak is blue and largely featureless, and its peak luminosity M_g = -21.0 places it with the most luminous members of the class. Multi-band photometry from ZTF, ATLAS, Pan-STARRS, LOT, LT, and Swift, plus Gemini and NOT spectroscopy, show the transient fading at dg/dt≈0.3 mag $d^{-1}$ and remaining blue (g-r≈-0.3) throughout. The spectra contain only narrow host-galaxy emission lines at a common redshift z=0.2267±0.0002, offset by 1.1 arcsec from the host centre, with lines extending to the transient position. Comparing rest-frame ugri lightcurves, the colour evolution and inferred temperature track AT 2018cow and AT 2020xnd almost exactly, making it a near twin of AT 2018cow. The discovery also serves as a proof of concept for the FastFinder alert-stream annotator, which flagged the source on the basis of its decline rate and blue colour.","pith_inferences":["If the near-identical match to AT 2018cow extends to X-ray and radio wavelengths, then models proposed for AT 2018cow—such as a central engine or a tidal disruption event—would apply to AT 2024qfm as well, strengthening the case for a common physical origin across the class.","The host's relatively old stellar population (mass-weighted age about 7 Gyr) and its projected offset of about 4 kpc suggest LFBOT progenitors can arise from older, lower-mass environments than typical core-collapse supernovae, potentially widening the search for progenitor channels.","Optimising the FastFinder thresholds and incorporating a second-epoch confirmation directly into the alert stream could cut the current five-day latency between first detection and spectroscopic classification, improving the odds of catching the next such event early enough for multi-wavelength campaigns."],"forward_implications":["Adds a fully characterised LFBOT at z=0.2267 to the still-small sample, confirming that LFBOT host offsets span a wide range and that host properties can include an old stellar population.","Demonstrates that a simple flux-gradient plus colour filter on a public alert stream can recover fast-evolving transients in real time, a method directly transferable to LSST.","If AT 2024kth is confirmed spectroscopically as an LFBOT, the two FastFinder discoveries would raise the ZTF-derived LFBOT rate by about 66 percent.","With LSST's depth, the same approach could detect LFBOTs out to z≈0.8, a survey volume roughly ten times larger than ZTF's, potentially finding about one per month.","The main bottleneck is latency: spectroscopic confirmation and multi-wavelength follow-up must happen within days, which will be challenging with LSST's inter-night cadence."],"supporting_citations":[{"why":"Supplies the AT 2018cow discovery spectrum and lightcurve that AT 2024qfm is directly compared against.","marker":"Prentice et al. 2018"},{"why":"Establishes the multi-wavelength definition of LFBOTs from AT 2018cow follow-up, including the X-ray and radio properties.","marker":"Perley et al. 2019"},{"why":"Provides the AT 2020xnd comparison LFBOT and its rest-frame lightcurve used in the match.","marker":"Perley et al. 2021"},{"why":"Defines the host-galaxy properties and offsets of the broader LFBOT sample used to contextualise AT 2024qfm's host.","marker":"Sevilla et al. 2026"},{"why":"Provides AT 2024wpp comparison data and the ZTF LFBOT rate that AT 2024qfm and AT 2024kth update.","marker":"Perley et al. 2026"},{"why":"Describes the Lasair broker infrastructure on which the FastFinder annotator runs.","marker":"Williams et al. 2024"},{"why":"Details the FastFinder annotator algorithm that flagged AT 2024qfm and AT 2024kth.","marker":"Fulton 2026"},{"why":"Supplies the forced photometry service from which the ZTF lightcurve points and decline rate are measured.","marker":"Masci et al. 2023"}],"fun_headline_variants":["AT 2024qfm: a fast blue transient, twin of AT 2018cow","New LFBOT AT 2024qfm is near twin of AT 2018cow","FastFinder filter snags AT 2024qfm, twin of AT 2018cow","Fast blue transient AT 2024qfm echoes AT 2018cow at z=0.2267","AT 2024qfm: luminous fast blue transient like AT 2018cow"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The transient is assumed to be at the redshift of its host galaxy (z=0.2267), derived from narrow nebular emission lines that extend to the transient position; if AT 2024qfm is not physically associated with this host, its derived absolute magnitude and LFBOT classification would be invalid.","fun_headline_variants_meta":{"raw":{"variants":["AT 2024qfm: a fast blue transient, twin of AT 2018cow","New LFBOT AT 2024qfm is near twin of AT 2018cow","FastFinder filter snags AT 2024qfm, twin of AT 2018cow","Fast blue transient AT 2024qfm echoes AT 2018cow at z=0.2267","AT 2024qfm: luminous fast blue transient like AT 2018cow"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000313,"raw_usage":{"total_tokens":1861,"prompt_tokens":1109,"completion_tokens":752,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":725,"completion_tokens_details":{"reasoning_tokens":625}},"tokens_in":725,"tokens_out":752,"duration_ms":6984,"temperature":1.0,"reasoning_tokens":625,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T18:45:26.298303+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a spatially resolved spectrum of the host across the transient position: if the narrow nebular lines at the transient are not at the same systemic velocity as the host, or come from a background galaxy, the redshift association—and with it the absolute magnitude and LFBOT classification—breaks.","supporting_citations":[],"review_version":1}