{"id":"bf69d33e-75cb-4d6f-b18b-eb1aeba892af","arxiv_id":"2608.10832","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Broad Balmer and Paschen line ratios in little red dots imply that some of these objects are seen through dust associated with the broad-line region, beyond the effects of dense optically thick gas.","lead":"Using JWST and SDSS/DESI spectra, the authors find that broad hydrogen line ratios in 20 little red dots are far above standard Case B values, and that several objects cannot be explained by dense, dust-free gas alone. They argue that dust sits in front of the broad-line region, not the narrow-line region, acting like a thin proto-torus around the central black hole.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Dust-free H-alpha/H-beta ceiling is convention-dependent; outward-emission models reach >100, so the 'at least half' population claim is not established.","rationale":"The reader correctly identifies the total-versus-outward Cloudy output convention as the weakest assumption. My reading of the text and Appendix B confirms that this choice is load-bearing specifically for the population-frequency claim: the ten sources with H-alpha/H-beta >13 but no Paschen lines are classified as dust-required solely because the chosen total-emission grid caps at ~13, whereas the outward-only dust-free grid reaches >100. The paper is transparent about this in Appendix B, and it argues that outward-only models do not simultaneously reproduce the joint Balmer-Paschen positions of the objects with Paschen detections. That argument can preserve the inference for J1022+0841 and the Rosetta Stone, but it does not rescue the extrapolation to 'about half of the population'. The concrete test I propose directly separates these two levels of the claim. Because the reader's conditional verdict already captures this uncertainty, I do not recommend moving the verdict; the appropriate adjustment is to keep the conclusion conditional and temper the population-wide dust fraction claim until directional line-transfer predictions and a larger Paschen sample are available.","tokens_in":31706,"tokens_out":4224,"duration_ms":53405,"concrete_test":"Rerun the same Cloudy grid but use I_out = I_tot - I_in as the model line ratio and redraw Figures 3, 5, and 6. Count how many of the ten H-alpha/H-beta>13 LRDs fall below the new dust-free ceiling and are no longer classified as dust-required. Then check whether J1022+0841 and the Rosetta Stone remain outside the dust-free joint H-alpha/H-beta versus Pa-beta/Pa-gamma and Pa-alpha/Pa-beta versus Pa-beta/Pa-gamma regions. If the outward grid encompasses the high-Balmer objects but still excludes the two Paschen sources, the robust conclusion shrinks to dust in some LRDs rather than dust in at least half.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 5.1 and Figures 5-6 adopt Cloudy's total line intensity I_tot = I_in + I_out for a plane-parallel slab, giving a dust-free maximum H-alpha/H-beta ~13. Appendix B (Figures B1-B3) shows that using only the outward (shielded-face) component I_out = I_tot - I_in, dust-free models reach H-alpha/H-beta >100 near log n_H ~ 8. The population-wide claim in the Abstract and Section 5.1 counts two high-z and eight low-z LRDs with H-alpha/H-beta >13 as 'cannot be reproduced by non-Case B models', and therefore as requiring dust. These ten sources lack Paschen constraints, so the convention choice is the only evidence for the 'at least about half' statement. The paper's justification for I_tot is reproducing the L_bol-L_H-alpha scaling from Yanagisawa et al. (2026a), which calibrates a luminosity normalization rather than the directional line-ratio convention. That does not establish that an observer would see summed two-sided emission instead of the directionally escaping component. The joint Balmer-Paschen evidence for J1022+0841 and the Rosetta Stone may survive under I_out, but the broader population claim is not secured.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper measures broad Balmer and, where available, Paschen hydrogen line ratios in 20 little red dots (LRDs), combining five high-redshift JWST/NIRSpec sources with fifteen low-redshift sources from the literature. Using Cloudy plane-parallel models with the total (inward plus outward) line emission, the authors find that the dust-free non-Case B grid can reach Hα/Hβ at most ~13, and that two of the three LRDs with reliable broad Paschen measurements (J1022+0841 and the Rosetta Stone) require additional dust extinction E(B−V) ≳ 0.2–1.0 even after accounting for high-density, optically thick effects. They further argue that, because two high-z and eight low-z LRDs have Hα/Hβ > 13, dust obscuration may be common in the LRD population, potentially representing a low-column counterpart of the AGN torus.","tokens_in":31914,"tokens_out":10878,"duration_ms":112120,"significance":"If the central inference holds, this is one of the first joint Balmer–Paschen demonstrations that non-Case B physics is active in LRD broad-line regions and that some LRDs require dust that is spatially associated with the broad-line gas, with implications for the weak hot-dust emission and for torus-like geometries. The measurement side is carefully executed: the paper uses MCMC posteriors, an absorption-component treatment with a ΔBIC criterion, a wavelength-dependent LSF, and an external, non-fitted Cloudy grid, which are strengths. However, the population-level conclusion depends critically on the adopted line-output convention, and the paper itself acknowledges in Appendix B that the Balmer decrement alone cannot exclude dust-free outward-only models; this convention dependence is not fully resolved for the Balmer-only subsample.","major_comments":[{"comment":"The central population claim—that LRDs with Hα/Hβ > 13 'cannot be reproduced' by dust-free non-Case B models—is convention-dependent. Appendix B explicitly shows that the outward-only (shielded-face) component reaches Hα/Hβ > 100 near log n_H ~ 8 and states that 'the Balmer decrement alone might not exclude a dust-free configuration dominated by shielded-face outward emission.' For an external observer of a spherical or quasi-spherical distribution of optically thick BLR clouds, the emergent radiation is the outward component, not the sum of inward and outward components; the paper's justification for adopting I_tot via the L_bol–L_Hα scaling (H. Yanagisawa et al. 2026a) constrains the total luminosity normalization, not the directional line-ratio convention. Therefore, the ten sources without Paschen constraints do not establish the abstract's statement that dust is present in at least about half of the LRD population, and the claim should be explicitly conditioned on the adopted geometry or supported with a direct test of the outward-only convention.","section":"Section 3.2, Appendix B, Section 5.1, Abstract"},{"comment":"The robustness of the joint Balmer–Paschen dust detection for the Rosetta Stone to the line-output convention is not demonstrated. The Rosetta Stone has Hα/Hβ ≈ 13.9, only slightly above the total-emission ceiling, and Paβ/Paγ ≈ 2.2. The paper only shows that J1022+0841's Paβ/Paγ remains outside the dust-free range under the outward-only prescription; it does not present the outward-only grid on the Hα/Hβ–Paβ/Paγ plane or evaluate the Rosetta Stone under that convention. If the Rosetta Stone can be reproduced without dust in the outward-only prescription, the number of robust dust detections reduces from two to one, materially weakening the central conclusion; this should be checked explicitly and the abstract and summary revised accordingly.","section":"Section 5.1, Summary item 4, Appendix B"}],"minor_comments":[{"comment":"The phrase 'at least about half of the population' should be replaced by 'about half of the present sample' unless a statistical completeness argument is supplied, since Section 2 states that the sample is not intended to be statistically complete.","section":"Abstract and Section 5.1"},{"comment":"Adding a second horizontal line or shaded band showing the outward-only dust-free maximum would make the convention dependence of the Hα/Hβ > 13 threshold visible to the reader.","section":"Figure 3"},{"comment":"When justifying the I_tot convention by the L_bol–L_Hα relation, state quantitatively how well the total-emission models reproduce that relation and note that a luminosity normalization provides no direct constraint on the line-ratio convention.","section":"Section 3.2"},{"comment":"The comparison between the sphere-total and plane-parallel-total models is made only by visual inspection of Figure B4; a quantitative measure of the agreement (for example, the median offset or the fraction of grid points within a given ratio tolerance) would strengthen the claim that the two geometries give broadly similar results.","section":"Appendix B"},{"comment":"The term 'proto-torus' is evocative and could be misread as implying an evolutionary connection to classical AGN tori; a footnote or explicit sentence clarifying that it is a purely descriptive term for a low-column dusty structure would avoid overinterpretation.","section":"Section 5.4"},{"comment":"There is a typo: 'maixmum' should be 'maximum' in the sentence following Figure 7.","section":"Section 5.1"}],"recommendation":"major_revision","confidential_remarks":"The measurement paper is technically careful, but the main population claim rests on the total-emission convention, which is arguably not the on-sky observable for a spherical BLR. I would encourage the authors to either adopt the outward-only convention as the primary reference for line-ratio comparisons, or to explicitly present all population statements as conditional on the total-emission geometry. The joint Paschen analysis for J1022+0841 appears robust, but the Rosetta Stone case needs a direct outward-only test before the 'two of three' conclusion is secure."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The real news here is the joint Balmer–Paschen analysis. Prior work on little red dots mostly used Balmer lines alone, so the Pa-alpha/Pa-beta ratios below Case B and the joint position of J1022+0841 are genuinely new. The line measurements look careful, with MCMC uncertainties and a reasonable treatment of absorption and neighboring lines. The conclusion that at least one LRD requires dust even if you allow outward-only emission is solid; that evidence does not depend on the geometry convention.\n\nThe soft spot is the population claim. The abstract and Section 5.1 say that H-alpha/H-beta > 13 in ten sources \"cannot be reproduced\" by dust-free models and therefore dust is common. That ceiling of 13 comes from the total two-sided Cloudy output. Appendix B shows that outward-only models reach H-alpha/H-beta above 100 near log n_H ~ 8. If those models are physically acceptable, sources with ratios between 13 and 100 do not require dust at all. The justification for using the total emission is that it reproduces the L_bol–L_H-alpha scaling from the same group's earlier paper, but that calibrates a luminosity normalization, not the directional line-ratio convention. So the \"at least about half\" statement is not secured. The paper is honest about this in Appendix B, but the abstract carries the stronger claim without the caveat.\n\nThe proto-torus picture is clearly labeled as illustrative, and the MIR comparison is appropriately qualified. The narrow-versus-broad line ratio contrast is a nice observational point on its own.\n\nWho gets value: anyone working on LRD central engines, broad-line physics, or AGN unification. The paper deserves a serious referee because the measurements and the joint diagnostic are worth publishing even if the population claim needs to be tempered. My recommendation: send it to review, but ask the authors to either defend the total-emission convention with a more physical argument or soften the abstract and Section 5.1 to match what the data actually show.","headline":"Careful line measurements and a genuinely new Balmer-Paschen diagnostic, but the population-wide dust claim rests on a convention-dependent ceiling that the paper itself shows is fragile.","tokens_in":32570,"tokens_out":1511,"would_cite":true,"duration_ms":19007,"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":"The paper claims that dust in the broad-line region of little red dots, revealed by combined Balmer and Paschen line ratios, is common and may be a low-column counterpart of AGN tori.","keywords":["little red dots","broad-line region","Balmer decrement","Paschen lines","dust attenuation","non-Case B recombination","photoionization modeling","active galactic nuclei"],"falsifier":"Measure the broad $\\mathrm{Pa}\\beta/\\mathrm{Pa}\\gamma$ and $\\mathrm{Pa}\\alpha/\\mathrm{Pa}\\beta$ ratios for the eight low-redshift LRDs with broad $H\\alpha/H\\beta>13$; if those ratios fall on the dust-free high-density model tracks, the dust inference for those objects fails. A single LRD with broad $H\\alpha/H\\beta$ of 5--10 and $\\mathrm{Pa}\\alpha/\\mathrm{Pa}\\beta$ below Case B, with narrow ratios at Case B, would confirm the non-Case B channel independent of dust.","tokens_in":31426,"feed_emoji":"🔴","tokens_out":11275,"duration_ms":102399,"temperature":0.7,"pith_summary":"Little red dots are compact, V-shaped objects whose broad Balmer lines often show $H\\alpha/H\\beta$ ratios far above the Case B recombination value. This paper assembles 20 LRDs with broad H $\\beta$ detected above $5\\sigma$, measures broad Balmer and, where available, Paschen lines, and compares them with dust-free photoionization grids. It finds that dense, optically thick gas alone can push $H\\alpha/H\\beta$ to about 13, but several objects exceed this, and the joint Balmer-Paschen fits require extra dust extinction of $E(B-V)\\sim0.2$--$1.0$ toward the broad-line region. Because the narrow-line ratios stay near Case B, the dust must be associated with the broad-line region rather than the galaxy-scale interstellar medium. If correct, dust is common around LRD nuclei and may be a low-column-density counterpart of the AGN torus.","feed_headline":"Dust around most little red dots veils the broad-line region","feed_subtitle":"Balmer and Paschen line ratios exceed dust-free dense-gas models, requiring circumnuclear dust in most little red dots.","key_machinery":"The central tool is the joint plane of broad $H\\alpha/H\\beta$ versus $\\mathrm{Pa}\\beta/\\mathrm{Pa}\\gamma$ (and $\\mathrm{Pa}\\alpha/\\mathrm{Pa}\\beta$), compared with a grid of constant-density, plane-parallel Cloudy models computed with the 'total' line intensity (inward plus outward emission). The non-Case B mechanism is the combined effect of large line optical depths, radiative trapping, and collisional excitation at $\\log n_{\\rm H}\\sim10$--$11\\,{\\rm cm^{-3}}$ and $\\log \\Phi({\\rm H})\\sim18\\,{\\rm photons\\,cm^{-2}\\,s^{-1}}$, which can raise $H\\alpha/H\\beta$ to about 13. The total-emission convention sets the dust-free ceiling; choosing only outward emission would raise the ceiling above 100 and remove the need for dust, which is why the convention is the load-bearing choice.","core_discovery":"The discovery claim is that the broad hydrogen line ratios of LRDs occupy a regime that dust-free non-Case B photoionization models cannot cover. Using plane-parallel Cloudy models with total two-sided line emission, the maximum dust-free $H\\alpha/H\\beta$ is about 13, reached near $\\log n_{\\rm H}\\sim10\\,{\\rm cm^{-3}}$ and $\\log \\Phi({\\rm H})\\sim18\\,{\\rm photons\\,cm^{-2}\\,s^{-1}}$. Of the three LRDs with reliable broad Paschen lines, one (J1047+0739) is reproduced by dense, high-density gas with $\\log n_{\\rm H}\\sim10$--$11$; the other two (J1022+0841 and the Rosetta Stone) require additional attenuation $E(B-V)\\sim0.2$--$1.0$ even after these non-Case B effects. Since their narrow $H\\alpha/H\\beta$ ratios are close to Case B, the extra extinction applies specifically to the broad-line component. In the full sample, two high-redshift and eight low-redshift LRDs have broad $H\\alpha/H\\beta>13$, the dust-free ceiling, implying that dust obscuration is present in at least about half of the population.","pith_inferences":["If the outward-only emission convention is the better description of what observers see, the dust-free ceiling rises far above 100 and the dust inference for objects only slightly above 13 disappears; this is testable by computing observer-facing line ratios from the same grids and comparing them with the same data.","The same Balmer-Paschen decomposition predicts a population of Type-2-like LRDs, objects with similar continua but no detectable broad lines, and gives a quantitative expectation for how often such objects should appear if the dust column depends on viewing angle.","The proposed proto-torus would create a correlation between broad $H\\alpha/H\\beta$ and rest-frame 3--5 $\\mu$m emission in a larger sample; measuring both for the same LRDs would test whether the attenuating dust is also the dust that glows in the mid-infrared.","Paschen-line spectroscopy of the eight low-redshift LRDs with $H\\alpha/H\\beta>13$ should be the quickest discriminator: if their $\\mathrm{Pa}\\beta/\\mathrm{Pa}\\gamma$ and $\\mathrm{Pa}\\alpha/\\mathrm{Pa}\\beta$ follow the high-density dust-free tracks, the dust interpretation for those objects would be wrong."],"forward_implications":["Broad $H\\alpha/H\\beta$ values above about 13 become a dust diagnostic: any LRD above this value needs attenuation toward its broad-line region, with $E(B-V)$ roughly 0.2 to 1.0.","The dust is not a galaxy-wide foreground screen, because narrow-line ratios remain close to Case B; it must sit close to the broad-line gas.","A substantial fraction of LRDs, at least about half of the studied sample, may contain a low-column-density dusty structure that behaves like a scaled-down AGN torus.","Weak hot-dust emission in many LRDs does not argue against nuclear dust, because modest column densities can produce the observed line ratios.","Objects like J1047+0739 show that some LRDs are consistent with dense dust-free gas, so the non-Case B channel must be included before attributing any Balmer excess to dust."],"supporting_citations":[{"why":"Supplies the three local LRDs with broad Balmer and Paschen fluxes, WISE photometry, and the dust-template decomposition used in the MIR comparison.","marker":"X. Lin et al. (2026a)"},{"why":"Provides the 12 DESI-selected local LRDs with broad Hα and Hβ fluxes used in the Balmer-decrement statistics.","marker":"X. Lin et al. (2026b)"},{"why":"Provides the high-redshift LRD compilation and the spectral selection from which the five high-z sources are drawn.","marker":"A. de Graaff et al. (2025a)"},{"why":"Identifies the Rosetta Stone and supplies its line and MIRI data used in the Balmer-Paschen plane.","marker":"I. Juodzbalis et al. (2024)"},{"why":"Reports RUBIES-BLAGN-1, the supplementary source with multiple Paschen lines but no broad Hβ coverage.","marker":"B. Wang et al. (2025)"},{"why":"Provides the Cloudy photoionization code used for the dust-free line-ratio grids.","marker":"G. J. Ferland et al. (1998)"},{"why":"Documents the Cloudy version c23.01 used in the calculations.","marker":"C. M. Gunasekera et al. (2023)"},{"why":"Supplies the L_bol-L_Halpha scaling relation that motivates the total-emission convention for the models.","marker":"H. Yanagisawa et al. (2026a)"},{"why":"Offers the outward-only model ratios that set an alternative dust-free ceiling, the main counterweight to the total-emission choice.","marker":"Z. Yan et al. (2026)"},{"why":"Provides the Case B recombination line ratios used as the baseline reference.","marker":"P. J. Storey & D. G. Hummer (1995)"}],"fun_headline_variants":["Most little red dots hide their broad-line regions in dust","Little red dots' extreme line ratios point to dusty broad-line regions","Dusty BLRs common in little red dots, new line ratios show","Little red dots' broad-line ratios demand dust, not just dense gas"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The dust requirement rests on the model convention that counts the total two-sided emission of a plane-parallel slab, which caps the dust-free $H\\alpha/H\\beta$ at about 13; counting only the outward-facing emission would let dust-free gas reach ratios above 100, removing the need for dust in objects with ratios between 13 and 100.","fun_headline_variants_meta":{"raw":{"variants":["Most little red dots hide their broad-line regions in dust","Little red dots' extreme line ratios point to dusty broad-line regions","Dusty BLRs common in little red dots, new line ratios show","Little red dots' broad-line ratios demand dust, not just dense gas"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000866,"raw_usage":{"total_tokens":3899,"prompt_tokens":1234,"completion_tokens":2665,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":850,"completion_tokens_details":{"reasoning_tokens":2588}},"tokens_in":850,"tokens_out":2665,"duration_ms":19920,"temperature":1.0,"reasoning_tokens":2588,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T16:25:55.051814+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the broad $\\mathrm{Pa}\\beta/\\mathrm{Pa}\\gamma$ and $\\mathrm{Pa}\\alpha/\\mathrm{Pa}\\beta$ ratios for the eight low-redshift LRDs with broad $H\\alpha/H\\beta>13$; if those ratios fall on the dust-free high-density model tracks, the dust inference for those objects fails. A single LRD with broad $H\\alpha/H\\beta$ of 5--10 and $\\mathrm{Pa}\\alpha/\\mathrm{Pa}\\beta$ below Case B, with narrow ratios at Case B, would confirm the non-Case B channel independent of dust.","supporting_citations":[],"review_version":1}