{"id":"05f277f7-ecfd-4e83-bddc-3ebd51d67088","arxiv_id":"2607.05182","paper_version":1,"verdict":"CONDITIONAL","confidence":"UNKNOWN","novelty_score":7.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":10,"one_line_summary":"Transient Faraday-complex spectropolarimetric structure detected during radio flaring of Swift J1727 implies internal Faraday rotation from electron-proton jet plasma with rotating mass ~10^21 g, a small fraction of accreted mass.","lead":"The paper uses MeerKAT radio spectropolarimetry to detect transient Faraday rotation structure in the jets of a black hole X-ray binary during its 2023 outburst, concluding the jets contain electron-proton plasma with a characteristic rotating mass of ~10^21 g. This matters because jet composition and mass loading are fundamental open questions in black hole astrophysics, and this introduces a new observational probe.","discovery_kind":"unclear","skeptic_critique":{"model":"glm-5.2","headline":"The internal-origin argument is the load-bearing premise; the disc-wind elimination is its weakest point but is more robust than a simple timescale objection suggests, because the transience of the complexity and the strong disfavoring of multi-thin-component models add independent support.","rationale":"The reader correctly identifies the internal-origin premise as load-bearing and the disc-wind timescale as its weakest point. I partially agree: the disc-wind argument is indeed the weakest link in the elimination chain, but the paper's case is stronger than the reader's framing suggests because (a) the transience of the complexity on ~2-day timescales is strong evidence against any slowly evolving external screen, and (b) the strong Bayesian disfavoring of multi-thin-component models (ΔlnZ > 10) rules out the most natural 'multiple sources through foreground' alternative.\n\nThe more specific concern I identify is that the mass estimate depends on Faraday-thick component widths that exceed the L-band maximum recoverable scale (W_max ~ 25 rad m⁻²). The paper acknowledges this and treats the mass as a lower limit, which is appropriate. However, the seed dependence of the October 14 fit and the broad posterior on the shape parameter N for several components suggest that the component decomposition itself may be unstable, not just the width. Since the mass estimate uses W_rm,99 from the October 06 fit (which is seed-stable), this concern is mitigated for the headline number, but the October 06 components still have widths (σ_φ ~ 21 and ~34 rad m⁻²) that approach or exceed W_max.\n\nThe composition argument is the most robust part of the paper. It follows directly from the physics of internal Faraday rotation (Eq. 6: φ_f ∝ q³m⁻², which cancels for charge-symmetric pair plasma) and is independent of the exact mass estimate. Even if the rotating material is from a previous ejection rather than the current flare, the composition conclusion holds as long as the material is jet-related.\n\nThe paper provides substantial independent support: multiple spectropolarimetric techniques (RM synthesis and QU-fitting) give consistent results; calibrator sources (3C286, J1733−1304) show stable behavior, ruling out instrumental systematics; P-component models are tested and disfavored; and the SSA-derived size scale is cross-checked against VLBI core-shift predictions (Eq. 24). The code and data are publicly available.\n\nThe CONDITIONAL verdict is appropriate. The observational result (transient Faraday complexity during flaring) is well-established. The composition argument is strong but contingent on the internal-origin premise, which is well-argued but not definitively proven without simultaneous VLBI spectropolarimetry. The mass estimate is explicitly exploratory and should be treated as an order-of-magnitude constraint. No verdict adjustment is needed.","tokens_in":63907,"tokens_out":9875,"duration_ms":227749,"concrete_test":"Inject a simulated Faraday-thick component with known parameters (e.g., σ_φ = 30 rad m⁻², N = 5, φ_rm = 50 rad m⁻², p₀ = 1%) into the October 06 epoch's actual λ² sampling, add realistic noise, and recover it with the same QU-fitting pipeline. Compare the recovered W_rm,99 to the injected value. If the recovered width deviates by more than 30%, the L-band coverage is significantly biasing the Faraday-thickness measurement, and the mass estimate M_rot ∝ W_rm,99 is unreliable at the factor-of-few level claimed. This can be done with existing data and the published codebase.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central claim — that the Faraday complexity is internal to the jet ejecta, implying electron-proton composition and enabling the mass estimate — rests on eliminating external-screen explanations. The ISM elimination is strong (stable foreground RM at −1.0±0.3 rad m⁻² across epochs and across ~5\" angular scales). The disc-wind elimination (Section 4.1) is weaker: it assumes a wind speed of ~1000 km/s and requires the wind to reach the ~10^15 cm emission radius, yielding a ~100 d travel time that exceeds the ~20 d interval from outburst onset to the first Faraday-thick epoch. However, faster winds (3000–5000 km/s, observed in some XRB systems) could reach 10^15 cm in ~20–30 days, and the paper's counter-argument that a faster wind would be less dense for fixed mass-loss rate does not account for the possibility of a higher mass-loss rate or a wind launched earlier in the outburst.\n\nThat said, the paper's argument is more robust than this single objection implies, for two reasons. First, the transience of the Faraday complexity — appearing and disappearing on ~2-day timescales coincident with flaring — is difficult to explain with any static or slowly evolving external screen. Second, the paper tests and strongly disfavors (ΔlnZ > 10) models with multiple Faraday-thin components (SS, SSS, etc.) and spectral power-law components (PP, PPP), which are the natural signatures of multiple unresolved sources viewed through a foreground screen. This makes a 'multiple components through a turbulent foreground' alternative less viable.\n\nThe quantitative mass estimate (M_rot ~ 10^21 g, f_rot ~ 10^-3) is more fragile. It depends on the characteristic Faraday thickness W_rm,99 ~ 100 rad m⁻², which is derived from components whose widths (σ_φ up to ~34 rad m⁻², W_rm,99 up to ~130 rad m⁻²) exceed the L-band maximum recoverable scale W_max ~ 25 rad m⁻². The paper acknowledges this and treats the mass as a lower limit, but the L-band filtering could bias not just the width","agreement_with_reader":"partial"},"referee_report":{"model":"glm-5.2","summary":"This paper presents MeerKAT L-band spectropolarimetry of the black hole X-ray binary Swift J1727.8-1613 during its 2023 outburst, focusing on the brightest radio flaring interval. Using both RM synthesis and parametric QU-fitting, the authors identify transient Faraday-complex structure (requiring Faraday-thick components) that appears and disappears on ~2-day timescales coincident with major radio flares. The stability of the foreground ISM rotation measure (-1.0 +/- 0.3 rad m^-2) across epochs and angular scales, together with the transience of the complexity and the strong disfavoring of multi-thin-component and spectral power-law models, is used to argue that the Faraday-rotating material is internal to the jet ejecta rather than in the ISM or a disc wind. Since internal Faraday rotation cancels in a charge-symmetric pair plasma, the authors conclude the ejecta must contain a dominant electron-proton component. Combining the characteristic Faraday thickness (W_rm,99 ~ 100 rad m^-2) with synchrotron self-absorption modeling of the Stokes I flare (yielding B_perp ~ 200 mG and l ~ 1.3x10^14 cm), they infer a Faraday-rotating mass M_rot ~ 10^21 g, corresponding to f_rot ~ 10^-3 of the accreted mass available during the flare. The paper also discusses the implications for electron acceleration efficiency and intrinsic polarisation-angle evolution.","tokens_in":64700,"tokens_out":1573,"duration_ms":163173,"significance":"The paper makes a timely and potentially impactful contribution by demonstrating that time-domain spectropolarimetry can be used as a diagnostic of jet composition and mass loading in transient relativistic jets, an approach that has been largely limited to AGN studies. The observational result -- transient Faraday complexity during flaring -- is supported by multiple independent methods (RM synthesis and parametric QU-fitting), conservative Bayesian evidence thresholds (Delta ln Z > 10), calibrator stability checks (Appendix C), and seed-dependence tests (Figure 10). The reproducible code and data products (GitHub repository, SSA codebase) are a notable strength. The argument for electron-proton composition, while resting on the interpretation of the Faraday complexity as internal, is physically well-motivated and supported by complementary mass-budget arguments from the literature. The mass estimate is appropriately framed as an order-of-magnitude exploratory calculation with clearly stated caveats. The falsifiable prediction that future outbursts should show similar transient Faraday-thick structure during flaring, and that broader-band spectropolarimetry should track the inward","major_comments":[{"comment":"Section 4.1, disc-wind elimination: The argument against a disc-wind origin for the Faraday complexity relies on a flow timescale of ~100 d for a ~1000 km/s wind to reach 10^15 cm, exceeding the ~20 d interval from outburst onset to the first Faraday-thick epoch. However, the counter-argument that a faster wind (3000-5000 km/s, observed in some XRB systems) would be less dense for fixed mass-loss rate does not account for the possibility of a higher mass-loss rate or a wind launched earlier in the outburst. The paper should more explicitly address whether a faster, denser wind launched during the hard-state rise could reach the relevant radius in time, or provide a stronger quantitative bound on the required mass-loss rate as a function of wind speed. That said, the transience of the complexity (~2-day timescales) and the strong disfavoring of multi-thin-component models (Delta ln Z > 10","section":null},{"comment":"Section 4.3, Eqs. (21)-(23): The mass estimate M_rot depends on W_rm,99 (fitted from polarimetric data), B_perp and l (from SSA modeling of Stokes I), and geometric/depolarization factors (D, theta_B). While these are derived from independent observables (polarized vs total intensity), the SSA model assumes a homogeneous emitting region, which the authors acknowledge is likely violated given the strong depolarization and inhomogeneous flare spectra. The sensitivity of M_rot to this assumption should be more explicitly quantified. Specifically, if the emitting region is inhomogeneous (as suggested by the sub-canonical optically thick spectral index), how much could B_perp and l change, and what is the resulting range on M_rot? The current treatment (Section 4.3.2) discusses this qualitatively but does not propagate the inhomogeneity into the mass uncertainty. This is load-bearing for the","section":null}],"minor_comments":[{"comment":"Section 3.3.3: The characteristic Faraday thickness W_rm,99 ~ 100 rad m^-2 is adopted from the flare-peak components, but the inferred widths range from 25 to 130 rad m^-2 across components and epochs (Section 3.3.3). The paper should clarify whether the mass estimate is sensitive to this choice, and whether using a different epoch (e.g., the more stable October 06 fit vs. the seed-dependent October 14 fit) would yield a substantially different M_rot.","section":null},{"comment":"Table 1: The table is very large and spans multiple pages. Consider splitting it or moving the full version to an appendix, showing only the favored models in the main text.","section":null},{"comment":"Section 4.3.1, Eq. (28): The angle theta_B between the ordered magnetic field and the line of sight is given a broad prior (10-80 deg, flat in cos theta_B). The resulting <B_parallel> ~ 30 mG has a very asymmetric uncertainty (+40, -20 mG). It would help to show the posterior distribution for <B_parallel> and M_rot, or at least state the median and 68% HDI more explicitly, to clarify how the mass estimate depends on this geometric factor.","section":null},{"comment":"Section 4.1.2: The approaching-receding ejecta interpretation for the paired thick components is interesting but speculative given the lack of VLBI evidence for a receding component. The Monte Carlo calculation showing R_F > 2 in ~20% of samples is useful, but the text could note more clearly that this is a consistency check rather than supporting evidence for this specific geometry.","section":null},{"comment":"Figure 4, middle panel: The green region showing a linearly evolving ISM contribution is mentioned but not clearly defined. A brief note on how this region is computed would help the reader.","section":null},{"comment":"Section 2.1.1: The manual correction of the cross-hand phase discontinuity (adding/subtracting pi beyond zero crossings) for epochs with large ionospheric RM could introduce subtle systematics. While Appendix C shows calibrator stability, a brief note on how this was validated for the target epochs would strengthen confidence in the spectropolarimetric fidelity.","section":null},{"comment":"The paper cites several works as submitted","section":null}],"recommendation":"minor_revision","confidential_remarks":"The paper is well-suited for MNRAS. The central observational result is robust and the physical interpretation is appropriately caveated. The main risk is that the internal-origin argument, while well-supported overall, has a identifiable weak point in the disc-wind elimination that the authors should address more thoroughly. The mass estimate is exploratory and should be treated as such, but it is clearly labeled as order-of-magnitude. I do not see grounds for rejection; the issues are addressable within the scope of a revision."},"author_rebuttal":{"model":"glm-5.2","summary":"We thank the referee for a careful and constructive report. Both major comments identify legitimate gaps in the quantitative treatment of systematic uncertainties. We address each below and describe the revisions we will make.","responses":[{"response":"The referee is correct that our treatment of the disc-wind scenario in Section 4.1 is incomplete. We considered only the case of a fixed mass-loss rate with varying wind speed, and did not quantitatively address the possibility of a higher mass-loss rate or a wind launched earlier in the outburst. We agree this gap should be closed. In the revised manuscript, we will add a quantitative bound on the required mass-loss rate as a function of wind speed and launch time. Specifically, we will compute the minimum wind mass-loss rate required to produce the observed Faraday thickness (W_rm,99 ~ 100 rad m^-2) at r ~ 10^15 cm, as a function of wind velocity (1000-5000 km/s) and launch epoch (ranging from outburst onset to ~10 days before the first Faraday-thick detection). This will allow the reader to directly compare the required mass-loss rates with observational constraints from Castro Segura et al. (2026) and typical XRB wind mass-loss rate estimates. We note that even in the most favorable case (5000 km/s wind launched at outburst onset), the required mass-loss rate likely exceeds the observationally inferred value by one to two orders of magnitude, but we will present this calculation explicitly rather than leaving it as a qualitative argument. We also agree that the transience of the complexity (~2-day timescales) and the strong disfavoring of multi-thin-component models provide independent constraints that are at least as important as the mass-budget argument, and we will make this clearer in the revised text.","revision_made":"yes","referee_comment":"Section 4.1, disc-wind elimination: The argument against a disc-wind origin for the Faraday complexity relies on a flow timescale of ~100 d for a ~1000 km/s wind to reach 10^15 cm, exceeding the ~20 d interval from outburst onset to the first Faraday-thick epoch. However, the counter-argument that a faster wind (3000-5000 km/s, observed in some XRB systems) would be less dense for fixed mass-loss rate does not account for the possibility of a higher mass-loss rate or a wind launched earlier in the outburst. The paper should more explicitly address whether a faster, denser wind launched during the hard-state rise could reach the relevant radius in time, or provide a stronger quantitative bound on the required mass-loss rate as a function of wind speed."},{"response":"The referee correctly identifies that the homogeneous SSA assumption is violated by the sub-canonical optically thick spectral index and strong depolarization, and that our current treatment of this systematic is only qualitative. We agree this should be quantified. In the revised manuscript, we will add a quantitative exploration of how inhomogeneity propagates into M_rot. Our approach will be as follows. The sub-canonical spectral index (alpha ~ +1.5 rather than +2.5 during the flare rise) can be modeled as a superposition of emitting regions with a distribution of optical depths (Jones & O'Dell 1977b; Cowie & Fender 2026). For a power-law distribution of component sizes or optical depths, the effective B_perp and l can differ from the homogeneous case by factors that depend on the breadth of the distribution. We will parameterize this using the inhomogeneous synchrotron model from Cowie & Fender (2026), which allows the spectral index to deviate from the homogeneous value, and propagate the resulting range of B_perp and l through to M_rot. Based on preliminary exploration, we expect the inhomogeneity to change B_perp by up to a factor of ~2-3 and l by a comparable factor, which would propagate into M_rot (which scales as B_perp^-1 * l^2) as an uncertainty of roughly one order of magnitude in either direction. This is consistent with our existing framing of M_rot as an order-of-magnitude estimate, but we agree it should be shown explicitly. We will add this as a subsection or extended discussion within Section 4.3.2, and will update the stated uncertainty range on M_rot accordingly. We note that the central conclusion -- that M_rot ~ 10^21 g represents a small fraction (~10^-3) of the accreted mass -- is robust to this uncertainty, since even an order-of-magnitude变化仍将","revision_made":"no","referee_comment":"Section 4.3, Eqs. (21)-(23): The mass estimate M_rot depends on W_rm,99 (fitted from polarimetric data), B_perp and l (from SSA modeling of Stokes I), and geometric/depolarization factors (D, theta_B). While these are derived from independent observables (polarized vs total intensity), the SSA model assumes a homogeneous emitting region, which the authors acknowledge is likely violated given the strong depolarization and inhomogeneous flare spectra. The sensitivity of M_rot to this assumption should be more explicitly quantified. Specifically, if the emitting region is inhomogeneous (as suggested by the sub-canonical optically thick spectral index), how much could B_perp and l change, and what is the resulting range on M_rot? The current treatment (Section 4.3.2) discusses this qualitatively but does not propagate the inhomogeneity into the mass uncertainty. This is load-bearing for the"}],"tokens_in":63948,"tokens_out":1161,"duration_ms":127600,"standing_objections":[]},"desk_editor":{"model":"glm-5.2","letter":"The headline result here is real: transient Faraday-complex spectropolarimetric structure appears during the radio flaring of Swift J1727 and disappears within days. This is the first time-domain application of these techniques to an XRB transient, and the observational detection is solid. The composition argument — that internal Faraday rotation requires an electron-proton component rather than a pure pair plasma — follows directly from the physics and is robust if the internal-origin premise holds. That's the part worth taking seriously, and it's genuinely new for XRB jets specifically. The paper also does the right things methodologically: multiple independent methods (RM synthesis + parametric QU-fitting), conservative evidence thresholds (ΔlnZ > 10), calibrator stability checks across the full campaign, seed-dependence tests, and public code and data on GitHub. The ISM elimination is strong — the foreground RM is stable at −1.0 ± 0.3 rad m⁻² across epochs and across ~5″ angular scales, which is hard to argue with. The disfavoring of multi-thin-component and spectral power-law alternatives is also convincing, particularly the point that P-models require unphysical β > 4 for the flare-peak epochs. The disc-wind elimination (Section 4.1) is the weakest link in the internal-origin argument, as the stress-test note correctly identifies. The ~1000 km/s wind speed and ~100 day travel time estimate could be challenged by faster winds or earlier wind launch. That said, the transience of the complexity — appearing and disappearing on ~2-day timescales coincident with flaring — is genuinely difficult to explain with any slowly evolving external screen, and this adds independent support that the stress-test note acknowledges. The internal-origin conclusion is probably right, even if the disc-wind argument alone isn't airtight. The quantitative mass estimate (M_rot ~ 10²¹ g, f_rot ~ 10⁻³) is more fragile and the authors know it. It depends on the SSA-derived B_perp ~ 200 mG and l ~ 1.3×10¹⁴ cm from a homogeneous emitting-region approximation that the strong depolarization and inhomogeneous flare spectra themselves contradict. Some fitted Faraday-thick widths (σ_φ up to ~34 rad m⁻², W_rm,99 up to ~130 rad m⁻²) exceed the L-band maximum recoverable scale of ~25 rad m⁻², which could bias not just the width but also the amplitude. The Oct 14 fit shows real seed dependence (Figure 10), though the Oct 6 fit is stable. The authors are appropriately cautious — they call the mass exploratory and a potential lower limit, and they identify the need for broader-band spectropolarimetry and simultaneous VLBI. This is a strong proof of concept. The qualitative composition result is the durable finding; the mass number is a first pass. The paper is for people working on jet physics, XRB outbursts, and anyone thinking about next-generation radio facilities. It deserves a serious referee who can engage with the spectropolarimetric methodology and the physical interpretation.","headline":"Letter re: arXiv:2607.05182","tokens_in":64919,"tokens_out":1467,"would_cite":false,"duration_ms":54892,"reading_group":"no","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["98.38.Mh","98.38.Fs","95.85.Bh"],"model":"glm-5.2","headline":"Polarised radio reveals proton-loaded jets from black hole binary","keywords":[],"falsifier":"Simultaneous VLBI spectropolarimetry showing no spatially resolved ejecta coincident with the Faraday-thick components would undermine the internal-rotation interpretation. Alternatively, if a future outburst shows Faraday-complex structure that does not correlate temporally with radio flaring, or if a demonstrated faster disc wind can reach the emission scale within the flare rise time, the external-screen hypothesis would regain traction and the composition inference would fail.","tokens_in":64141,"feed_emoji":"📡","tokens_out":1285,"duration_ms":118405,"temperature":0.7,"pith_summary":"This paper argues that transient, wavelength-dependent distortions in the polarised radio emission from the black hole X-ray binary Swift J1727.8–1613 are caused by Faraday rotation occurring *inside* the jet plasma itself, not in foreground material. Because internal Faraday rotation cancels out in a pure electron–positron plasma (equal numbers of positive and negative charges rotating the polarisation angle in opposite directions), detecting it requires a dominant electron–proton component in the jet ejecta. The authors identify transient Faraday-complex spectropolarimetric structure that appears and disappears in lockstep with the brightest radio flares, while the foreground rotation measure stays constant at about –1 rad m⁻². Using parametric QU-fitting and RM synthesis on MeerKAT L-band data, they model this structure as Faraday-thick components — emission mixed with rotating plasma — and argue against origins in the interstellar medium or a disc wind on temporal and geometric grounds. Anchoring the magnetic field (~200 mG) and size scale (~1.3×10¹⁴ cm) with synchrotron self-absorption modelling of the flare, they infer a characteristic Faraday-rotating mass of ~10²¹ g, roughly 10⁻³ of the mass accreted during the flare. They further show that the relativistic synchrotron-emitting electrons alone are too few and too inefficient as Faraday rotators to explain the observed rotation, requiring a colder electron–proton reservoir that dominates the plasma content but contributes little to the synchrotron emission. The paper also reports unusual evolution of the intrinsic polarisation angle, including post-flare obliquity and a 90-degree inversion between the unresolved core and resolved ejecta at late times, which may reflect relativistic beaming effects in helical magnetic fields.","feed_headline":"Polarised radio reveals proton-loaded jets from black hole binary","feed_subtitle":"Transient Faraday rotation inside jet ejecta of Swift J1727 requires electron–proton plasma, constraining jet mass loading to 10⁻³ of accret","key_machinery":"The load-bearing mechanism is the distinction between external and internal Faraday rotation. Faraday rotation scales as φ_f ∝ q³m⁻² ∫ n_e B_∥ dl, where the odd power of charge q means electrons and positrons contribute with opposite signs. In a charge-symmetric pair plasma, these contributions cancel exactly; in an electron–proton plasma, the proton contribution is suppressed by (m_e/m_p)² and electrons dominate. The paper models the observed spectropolarimetric complexity using super-Gaussian Faraday-thick components (following Anderson et al. 2016), where the shape parameter N interpolates between Gaussian external Faraday dispersion (N=2) and a Burn-slab top-hat (N>15). The mass estimate","core_discovery":"The central discovery is that transient Faraday-complex spectropolarimetric structure, appearing coincident with radio flares in a black hole X-ray binary, can be attributed to internal Faraday rotation within the jet ejecta. This attribution requires the jet plasma to contain a dominant electron–proton component rather than a pure electron–positron pair plasma, because internal rotation cancels in charge-symmetric pair plasmas. Combining the inferred Faraday thickness (~100 rad m⁻²) with synchrotron self-absorption estimates of the magnetic field and emitting region size yields a characteristic Faraday-rotating mass of order 10²¹ g — about 10⁻³ of the accreted mass available during the flar","pith_inferences":[],"forward_implications":["If internal Faraday rotation is a reliable diagnostic of jet composition, time-domain spectropolarimetry of X-ray binary flares can systematically distinguish electron–proton from electron–positron dominated jets across the transient jet population, something total-intensity monitoring cannot do.","The mass fraction f_rot ~ 10⁻³ inferred during flaring is much lower than the mass fractions (~1) inferred from deceleration modelling of resolved ejecta at late times, suggesting jets gain substantial mass through entrainment or environmental interaction as they propagate to larger scales.","The electron acceleration efficiency inferred from the ratio of relativistic to cold electron densities (≳10% by number) is higher than typical PIC simulation values for collisionless shocks (0.5–2%), hinting that particle acceleration may be especially efficient during the ejection phase, or that the cold electron reservoir is larger than estimated.","The polarisation-angle obliquity and 90-degree core–ejecta inversion could serve as a diagnostic of bulk relativistic speed if explained by aberration in helical magnetic fields, offering a complementary velocity probe for unresolved jets.","The method generalises to other transient jetted sources — tidal disruption events, gamma-ray bursts, and neutron star mergers — where jet composition and mass loading remain poorly constrained."],"fun_headline_variants":["Faraday rotation in black hole jet signals proton-loaded plasma","Radio polarization reveals baryonic mass in black hole binary jet","Protons dominate jet plasma in black hole X-ray binary flare","Transient radio flare polarization tracks baryonic jet mass loading","Internal Faraday rotation in relativistic jet requires protons"],"cache_read_input_tokens":0,"weakest_assumption_plain":"The interpretation depends on the observed Faraday complexity being internal to the jet ejecta rather than produced by a transient external screen local to the source. The paper argues against a disc-wind origin by estimating that a ~1000 km/s wind would need ~100 days to reach the relevant emission scale, but a faster wind or different geometry could change this conclusion. If the rotating material is instead external to the emitting plasma, the composition and mass in do","fun_headline_variants_meta":{"raw":{"variants":["Faraday rotation in black hole jet signals proton-loaded plasma","Radio polarization reveals baryonic mass in black hole binary jet","Protons dominate jet plasma in black hole X-ray binary flare","Transient radio flare polarization tracks baryonic jet mass loading","Internal Faraday rotation in relativistic jet requires protons","Jet mass loading traced by Faraday rotation in black hole binary","Spectropolarimetry exposes electron-proton jet in black hole binary","Baryonic mass in black hole jet measured via radio polarization"]},"model":"glm-5.2","effort":"high","cost_usd":0.0,"raw_usage":{"total_tokens":1354,"prompt_tokens":739,"completion_tokens":615,"prompt_tokens_details":null},"tokens_in":739,"tokens_out":615,"duration_ms":15810,"temperature":1.0,"reasoning_tokens":537,"cache_read_input_tokens":0,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-08T00:31:42.084997+00:00","model_set":{"reader":"glm-5.2"},"falsifier":"Simultaneous VLBI spectropolarimetry showing no spatially resolved ejecta coincident with the Faraday-thick components would undermine the internal-rotation interpretation. Alternatively, if a future outburst shows Faraday-complex structure that does not correlate temporally with radio flaring, or if a demonstrated faster disc wind can reach the emission scale within the flare rise time, the external-screen hypothesis would regain traction and the composition inference would fail.","supporting_citations":[],"review_version":1}