A weakly modelled view of the joint compact-binary mass plane: population structure and spectral-siren cosmology

As a fastidious and diligent researcher, I have meticulously analyzed both provided texts (A and B) pertaining to arXiv paper A, weakly modelled view of the joint compact-binary mass plane: population structure and spectral-siren cosmology.

The following is a comprehensive, detailed synthesis combining the core scientific findings from both sections to provide a thorough understanding of the paper's methodology, results, and conclusions.

Comprehensive Research Summary: Weakly Modelled View of Joint Compact-Binary Mass Plane

This research introduces a novel framework designed to simultaneously infer the joint source-frame component-mass distribution of compact binaries with cosmological parameters (specifically H 0), while providing a rigorous method for identifying and quantifying structures within the reconstructed density. The core innovation lies in employing a flexible mixture model for the mass distribution, coupled with an automated morphological identification framework that assesses posterior stability, local prominence, and cosmological relevance of these structures.

Methodology: The Flexible Joint Model

The central modeling approach utilizes a flexible mixture model to describe the joint source-frame component-mass distribution. This flexibility is crucial as it allows the model to emerge not just from a simple one-dimensional primary mass spectrum, but from the data itself, enabling the inference of both component masses and their relative distribution in two dimensions.

The framework incorporates an explicit mechanism for morphology classification, which moves beyond visual inspection of reconstructed densities or marginal distributions. Instead, structures are assigned quantitatively based on their two-dimensional position in the mass plane (m 1 vs m 2). This allows for a reproducible interpretation of familiar and emerging population structures.

Cosmological Constraints and Precision

The primary strength of this joint approach is its impact on cosmological constraints, particularly for H 0. The study demonstrates significant improvements in precision compared to existing methods:

• Narrower Constraint: The two-dimensional mass model constraint on H 0 is 33% narrower than analyses that flexibly reconstruct only the primary mass.

• Comparable Precision: It is only about 10% wider than analyses based on strongly parameterized one-dimensional spectral structures.

• Robustness to Model Choice: The near-LVK precision achieved by this joint reconstruction suggests that the additional information carried by the two-dimensional morphology can compensate for relaxing prescribed spectral structure, thereby reducing sensitivity to mass-spectrum model systematics while retaining comparable cosmological precision.

Structure Identification and Hierarchy

The framework successfully identifies and categorizes population structures, establishing a hierarchy based on their cosmological dependence:

1. Dominant Anchors (Strongest Dependence): The **C10 feature at (10.3, 8.6) M ** is identified as the dominant spectral-siren anchor. Its cosmological dependence is carried almost entirely by a coherent two-component mass scale (m). Similarly, the **C35 feature at (33.4, 29.5) M ** serves as the second main anchor, also dominated by this common joint mass scale.

2. Intermediate Structures (Residual Dependence): The weaker intermediate features and valleys exhibit greater sensitivity to population model freedom and the feature-finder threshold used for catalogue inclusion. These structures retain residual dependence in the orthogonal mass-ratio direction (q f), indicating that a single coherent rescaling of both component masses may not exhaust all cosmological information.

3. Weaker Anchors: Features like C20 (at 21.4, 11.8 M) and Vint show weaker overall dependence on H 0 but possess the clearest residual mass-ratio dependence, making them particularly sensitive to the two-dimensional geometry.

4. T60: This feature shows low posterior recovery regarding its joint morphology, although its global primary-mass reconstruction and H 0 posterior remain broadly stable.

Key Theoretical Insights

The analysis reveals a critical distinction in how cosmological dependence manifests across different features:

• Common Scale Dependence: Features like C10 and C35 are primarily governed by a common mass scale (m), which captures the entire (1+z) cosmological rescaling of both masses.

• Orthogonal Dependence: The intermediate unequal-mass feature retains residual dependence in the orthogonal mass-ratio direction (q f), suggesting that information exists in the joint mass plane that cannot be represented by a structured primary-mass spectrum combined with a global q beta prescription.

Robustness and Sensitivity Testing

To rigorously validate these claims, extensive controlled variations were performed on the model parameters:

• Model Complexity: Tests varied the mixture prior (mu alpha-1), the maximum number of components (K max from 10 to 100), and component width scales (L small).

• Prior Strength: Variations tested both weaker aspect priors (allowing more anisotropic components) and stronger aspect priors (disfavoring highly elongated components).

• Likelihood Treatment: The study explored alternative marginal likelihood treatments for low-mass ratio features (Vlow, Vint) and the treatment of likelihood uncertainty (sigma squared L < 1).

Results of Sensitivity Tests:

Crucially, the principal structures (Clow, C10, C35) remain identifiable across all population-model variations. The H 0 posteriors remain strongly overlapping across all tested configurations, with medians spanning 54.4–60.5 km s-1 Mpc-1, which is consistent with the baseline result (59.5 km s-1 Mpc-1). This stability confirms that the cosmological inference is robust against choices controlling mixture representation and likelihood treatment.

However, the weaker structures (C20 and Vint) are shown to be more sensitive to configurations that allow for additional small-scale and shape freedom. Furthermore, the residual test for mass-ratio dependence yields minimum raw p-values of 0.0025 for C20 and Vint, quantifying their sensitivity to this orthogonal dependence.

Conclusion

The paper concludes that explicit morphology classification transforms reconstructed densities into reproducible objects whose astrophysical and cosmological properties can be quantified across the posterior. This framework successfully navigates the trade-off between flexibility and precision: a weakly modelled reconstruction retains nearly the precision of strongly parameterized models while significantly reducing sensitivity to mass-model systematics. The findings strongly advocate for a genuinely two-dimensional view of compact binary populations, as this geometry is necessary to fully exploit spectral-siren information, especially concerning the residual mass-ratio dependence observed in intermediate structures. This flexible, two-dimensional approach represents a vital direction for future spectral-siren cosmology.

1. textbfWeighting of Population Features by Morphology Stability and Prominence: Improved Feature Classification and Interpretation Tools. The improved system can quantitatively distinguish between a genuine low-contrast feature may fall below a catalogue cut, while apparent weak structure may also arise from finite-catalogue fluctuations by explicitly calculating the morphology stability quantifies how frequently a compatible structure is recovered across posterior realizations, Pmorph,f and comparing it to Feature strength quantifies how strongly the recovered structure stands out from its local surroundings.

2. textbfDecomposition of Cosmological Dependence: Disentangling Common Scale vs. Residual Mass-Ratio Effects. The system can rigorously determine which structures act as spectral-siren anchors by decomposing their position into a common mass scale m∥ = √x1,fx2,f and the orthogonal mass-ratio coordinate qf = x2,f /x1,f to quantify whether coherent rescaling of the two component masses might not exhaust the cosmological dependence of every population structure.

3. textbfAdaptive Model Selection and Complexity Control: Optimizing Model Flexibility for Systematics Reduction. The improved system can dynamically adjust model parameters based on data quality by using diagnostics like effective mixture complexity, Keff = K / Xmax k=1 we 2k!−1 and selecting characteristic widths based on event-level resolution Lj = fj Qpj Nobs i=1, j ∈ (B8) to ensure the reconstruction is not driven by activating the full mixture dictionary nor of pushing the basis functions to the narrowest scale allowed by construction.

4. textbfHandling Ambiguity in Feature Classification: Resolving Projection-Dependent Interpretations. The system can resolve ambiguities where a feature's classification depends on projection, specifically clarifying that the ambiguity in interpretation of this feature in the one dimensional marginal as either a peak or a slope change is a property of the one-dimensional projection rather than of the joint population morphology.

We introduce a flexible mixture model for the joint source-frame component-mass distribution of compact binaries, inferred simultaneously with cosmology, together with a framework that identifies structures in the reconstructed density, quantifying their posterior stability, local prominence, and cosmological relevance. Applied to GWTC-5.0, the workflow clarifies several familiar and emergent structures. The about35 M feature is classified as a clear near-equal-mass peak despite an ambiguous interpretation in the primary-mass projection. A weaker unequal-mass concentration near (21,12) M is recovered frequently but with lower prominence. A tentative joint transition near 60 M appears at mildly unequal masses, with a more robust manifestation in the primary-mass projection. For cosmology, the two-dimensional mass model constraint on H 0 is 33% narrower than a model that flexibly reconstructs only the primary mass, and only about 10% wider than analyses based on strongly parameterized one-dimensional spectral structure. At the feature level, we show that the about10 M and about35 M concentrations depend on H 0 mainly through a common scale which captures the entire about (1+z) cosmological rescaling of both masses; the intermediate unequal-mass feature instead retains residual dependence in the orthogonal mass-ratio direction. More generally, we argue that population prominence, formal inclusion in an astrophysical feature catalogue, and cosmological relevance need not coincide. These results suggest that a genuinely two-dimensional view may be required to fully exploit spectral-siren information. Our workflow provides a reproducible way to detect population structure and identify which aspects of its geometry are associated with spectral-siren constraints. The feature finder is directly applicable to other population reconstructions.