pith. sign in

arxiv: 2509.20430 · v2 · pith:ZE3OYOBYnew · submitted 2025-09-24 · 🌌 astro-ph.GA · astro-ph.CO

pop-cosmos: Star formation over 12 Gyr from generative modelling of a deep infrared-selected galaxy catalogue

Sinan Deger , Hiranya V. Peiris , Stephen Thorp , Daniel J. Mortlock , Gurjeet Jagwani , Justin Alsing , Boris Leistedt , Joel Leja This is my paper

Pith reviewed 2026-05-18 13:59 UTC · model grok-4.3

classification 🌌 astro-ph.GA astro-ph.CO
keywords star formation rate densitygalaxy quenchinggenerative modelingstellar population synthesisCOSMOS surveyAGN feedbackcosmic star formation historynon-parametric star formation histories
0
0 comments X

The pith

A generative model trained on 420,000 infrared-selected galaxies shows the cosmic star formation rate density peaked at redshift 1.3 and maps distinct star-formation histories for star-forming versus quiescent systems.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper applies a score-based diffusion model called pop-cosmos to 26-band photometry of a deep IRAC-selected sample from COSMOS2020. It directly integrates recovered star-formation rates to trace the cosmic star-formation rate density back to redshift 3.5. The work then extracts non-parametric star-formation histories and separates star-forming from quiescent galaxies using specific star-formation rate. This separation reveals gradually decorrelating activity in star-forming galaxies and abrupt quenching transitions in quiescent ones. The model also links peak AGN infrared output to the moment massive galaxies cross from star-forming to quiescent states.

Core claim

The central claim is that a score-based diffusion generative model, trained to match the joint distribution of 16 stellar population synthesis parameters on 420,000 IRAC-selected COSMOS2020 galaxies, recovers individual star-formation rates sufficiently well to compute the star-formation rate density by direct summation. This yields a peak at z = 1.3 ± 0.1 of 0.08 ± 0.01 M⊙ yr⁻¹ Mpc⁻³. Non-parametric histories show star-forming galaxies with star-formation-rate correlations that fall from near unity to zero over 13 Gyr, while quiescent galaxies maintain full recent correlation followed by sharp decorrelation on roughly 1 Gyr timescales for systems above 10¹⁰ M⊙. Infrared AGN luminosity is at

What carries the argument

pop-cosmos, a score-based diffusion generative model that learns the joint posterior distribution over 16 stellar population synthesis parameters directly from 26-band photometry of an IRAC Ch.1 < 26 selected sample of 420,000 COSMOS2020 galaxies.

If this is right

  • The star-formation rate density reaches a maximum of 0.08 M⊙ yr⁻¹ Mpc⁻³ at redshift 1.3 and declines both toward the present and toward earlier epochs.
  • Star-forming galaxies exhibit star-formation-rate correlations that decrease smoothly from r ≈ 1 at recent times to r ≈ 0 at lookback times of 13 Gyr.
  • Quiescent galaxies display full correlation over the most recent 300 Myr followed by abrupt decorrelation, indicating quenching transitions on ~1 Gyr timescales for galaxies with stellar masses between 10¹⁰ and 10¹¹ M⊙.
  • Infrared AGN luminosity reaches its highest values in massive galaxies precisely as they approach the star-forming to quiescent transition and drops once quiescence is established.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If the recovered quenching timescale of ~1 Gyr holds in larger samples, semi-analytic models and hydrodynamical simulations will need to incorporate feedback mechanisms that operate on that specific duration rather than instantaneous or multi-Gyr processes.
  • The finding that AGN infrared output peaks only during the transition phase suggests future multi-wavelength surveys could test whether black-hole accretion is causally required for quenching by searching for similar timing offsets in independent AGN indicators.
  • Extending the same generative approach to wider-area or deeper infrared surveys could test whether the reported absence of low-mass quiescent galaxies at z ~ 1 persists or is an artifact of the current sample depth.

Load-bearing premise

The diffusion model recovers the true joint posterior over the 16 stellar population parameters without substantial selection biases or misspecification that would distort integrated star-formation rate densities or the derived correlations in star-formation histories.

What would settle it

An independent analysis of the same COSMOS2020 sample using spectroscopic redshifts and alternative SED-fitting codes that returns a star-formation rate density peak redshift differing by more than 0.3 or a quiescent fraction at z ~ 1 that differs by more than 15 percent would falsify the central results.

Figures

Figures reproduced from arXiv: 2509.20430 by Boris Leistedt, Daniel J. Mortlock, Gurjeet Jagwani, Hiranya V. Peiris, Joel Leja, Justin Alsing, Sinan Deger, Stephen Thorp.

Figure 1
Figure 1. Figure 1: The pop-cosmos cosmic SFRD in bins of width Δ𝑧 = 0.2. Dark- and light-red shaded regions show, respectively, our COSMOS2020- and model￾thresholded estimates. Black curve is from Madau & Dickinson (2014). Gray and orange curves are from Behroozi et al. (2019). Blue markers show the literature compilation assembled by Behroozi et al. (2019) and updated in this paper. Black and purple markers highlight result… view at source ↗
Figure 2
Figure 2. Figure 2: Rest-frame 𝑁𝑈𝑉𝑟 𝐽 colour–colour diagrams. Cells are shaded based on the median sSFR, and are only shaded when they contain 𝑁 > 5 galaxies. The orange line shows the SF/Q boundary from Weaver et al. (2023b). In our colour-based analysis, Q galaxies are those to the upper left of the boundary. The sSFR range on the colourbar is limited to [10−11 , 10−8.5 ] yr−1 . 0.0-0.5 0.5-0.8 0.8-1.0 1.0-1.1 1.1-1.3 1.3-1… view at source ↗
Figure 3
Figure 3. Figure 3: Left: Contaminant fraction – i.e. false discovery rate, FP/(TP + FP) – for Q galaxies identified via 𝑁𝑈𝑉𝑟 𝐽, in bins of redshift and stellar mass above our mass-completeness limit. Cells above the mass-completeness threshold are populated if they contain at least 1000 galaxies. Right: Same as left panel, but with bins shaded based on the median log10(sSFR) of the galaxies in the bin. Each redshift (column)… view at source ↗
Figure 4
Figure 4. Figure 4: Quenched fraction as a function of stellar mass in bins of redshift. Each redshift bin contains 10 percent of the sample. Left: SF/Q selection based on 𝑁𝑈𝑉𝑟 𝐽 diagram. Right: SF/Q selection based on an sSFR threshold of 10−11 yr−1 . 0 1 2 3 redshift 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 quenched fraction log10(sSFR) < 11.0 [yr 1 ] log10(M* / M ) [7.2, 8.8) [8.8, 9.0) [9.0, 9.2) [9.2, 9.3) [9.3, 9.5) [9.5, 9.7) [… view at source ↗
Figure 5
Figure 5. Figure 5: The quenched fraction as a function of redshift in bins of stellar mass. Each stellar mass bin contains 10 percent of the sample, except for the two most-massive bins, which contain 10 percent between them. Left: SF/Q selection based on 𝑁𝑈𝑉𝑟 𝐽 diagram. Right: SF/Q selection based on an sSFR threshold of 10−11 yr−1 . Estimated cosmic variance for a COSMOS-sized field is shown as a shaded region for the most… view at source ↗
Figure 6
Figure 6. Figure 6: The stellar mass distribution of the full (left), star-forming (middle), and quiescent galaxies from pop-cosmos. Cells are shaded by number of galaxies. Only galaxies above our mass completeness threshold (red curve, left panel) are included. Faisst et al. 2017; Ellison et al. 2024; Heckman et al. 2024), nor extremely gradual (e.g. starvation of gas; Bekki et al. 2002; Van Den Bosch et al. 2008; Peng et al… view at source ↗
Figure 7
Figure 7. Figure 7: The SMF of the full (left), star-forming (middle), and the quiescent galaxies (right) in bins of redshift. Shaded regions show uncertainty due to cosmic variance and Poisson noise. We overplot the Schechter functions from Weaver et al. (2023b), including their reported uncertainty due to Poisson noise, cosmic variance, and stellar mass uncertainty from SED fitting. Redshift bins are from Weaver et al. (202… view at source ↗
Figure 8
Figure 8. Figure 8: The redshift-binned SMF of the full (left), star-forming (middle), and quiescent (right) samples. Redshift bins contain 10 percent of the full sample. Shaded regions show uncertainty due to cosmic variance and Poisson noise. Only galaxies above our mass-completeness threshold (see Section 2.4) are included. Matharu et al. 2021; Alberts & Noble 2022). We observe a minor increase of 0.5 dex in the peak mass … view at source ↗
Figure 9
Figure 9. Figure 9: Star formation and stellar mass assembly history of the SF sample at 1.0 < 𝑧 < 1.4, with 10 < log10 (𝑀∗/M⊙ ) < 11 and log10 (sSFR/yr−1 ) > −11. Top left: Correlation matrix between the SFH bins of this sample. Colourbar range is fixed to [0, 1]. Top right: The SFR distributions in the 7 SFH bins. Bottom left: Median SFR and standard deviation per SFH bin for this sample (black), and representative SFHs for… view at source ↗
Figure 10
Figure 10. Figure 10: Same as [PITH_FULL_IMAGE:figures/full_fig_p014_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Distributions of SFR ratios between the SFH bins for all the galaxies in 1.0 < 𝑧 < 1.4. SFH bins 𝑛 = 1 and 𝑛 = 7, respectively correspond to the most recent and earliest epochs of the SFH. The SFR ratio plotted is therefore the ratio of a more recent bin to the immediately preceding bin. The dashed vertical line shows an SFR ratio of one. The black histogram shows the Student’s 𝑡 continuity prior from Lej… view at source ↗
Figure 12
Figure 12. Figure 12: Mass dependent contribution to the SFRD, with the same redshift binning as [PITH_FULL_IMAGE:figures/full_fig_p016_12.png] view at source ↗
Figure 13
Figure 13. Figure 13: Stellar mass doubling time-scale (inverse sSFR; see, e.g. Bower et al. 2017) vs. stellar mass in bins of redshift. Left: Cells shaded based on galaxy count. Right: Cells shaded based on median AGN bolometric luminosity fraction, 𝑓AGN. The dashed gray line indicates a doubling time of 100 Gyr, equivalent to our SF/Q boundary of sSFR = 10−11 yr−1 . The vertical dashed lines show the mass completeness limit … view at source ↗
read the original abstract

We study star formation over 12 Gyr using pop-cosmos, a generative model trained on 26-band photometry of 420,000 COSMOS2020 galaxies (IRAC Ch.1 $<26$). The model learns distributions over 16 SPS parameters via score-based diffusion, matching observed colours and magnitudes. We compute the star formation rate density (SFRD) to $z=3.5$ by directly integrating individual galaxy SFRs. The SFRD peaks at $z=1.3\pm0.1$, with peak value $0.08\pm0.01$ M$_{\odot}$ yr$^{-1}$ Mpc$^{-3}$. We classify star-forming (SF) and quiescent (Q) galaxies using specific SFR $<10^{-11}$ yr$^{-1}$, comparing with $NUVrJ$ colour selection. The sSFR criterion yields up to 20% smaller quiescent fractions across $0<z<3.5$, with $NUVrJ$-selected samples contaminated by galaxies with sSFR up to $10^{-9}$ yr$^{-1}$. Our sSFR-selected stellar mass function shows a negligible number density of low-mass ($<10^{9.5}$ M$_\odot$) Q galaxies at $z\sim1$, where colour-selection shows a prominent increase. Non-parametric star formation histories around the SFRD peak reveal distinct patterns: SF galaxies show gradually decreasing SFR correlations with lookback time ($r\sim1$ to $r\sim0$ over 13 Gyr), implying increasingly stochastic star formation toward early epochs. Q galaxies exhibit full correlation ($r>0.95$) during the most recent $\sim$300 Myr, then sharp decorrelation with earlier star-forming epochs, marking clear quenching transitions. Massive ($10<\log_{10}(M_*/$M$_{\odot})<11$) galaxies quench on a time-scale of $\sim1$ Gyr, with mass assembly concentrated in their first 3.5 Gyr. Finally, AGN activity (infrared luminosity) peaks as massive ($\sim10^{10.5}$ M$_\odot$) galaxies approach the transition between star-forming and quiescent states, declining sharply once quiescence is established. This provides evidence that AGN feedback operates in a critical regime during the $\sim1$ Gyr quenching transition.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript introduces pop-cosmos, a score-based diffusion generative model trained on 26-band photometry for 420,000 IRAC Ch.1<26 selected galaxies from COSMOS2020. The model learns the joint distribution over 16 SPS parameters. SFRD is computed by direct integration of individual galaxy SFRs drawn from the model, yielding a peak at z=1.3±0.1 with value 0.08±0.01 M⊙ yr⁻¹ Mpc⁻³. Galaxies are classified as star-forming or quiescent using an sSFR<10^{-11} yr^{-1} threshold (compared to NUVrJ colour selection), revealing up to 20% differences in quiescent fractions and a negligible low-mass quiescent population at z~1. Non-parametric SFHs are analysed via correlation matrices, showing gradual decorrelation for SF galaxies and sharp ~1 Gyr quenching transitions for Q galaxies. AGN IR luminosity is shown to peak as massive galaxies approach the SF-to-Q transition.

Significance. If the diffusion model recovers unbiased joint posteriors, the work provides a statistically powerful, data-driven view of star formation over 12 Gyr with concrete, falsifiable results on SFRD evolution, SFH stochasticity, quenching timescales, and AGN feedback timing. The direct integration approach and correlation-matrix analysis of non-parametric histories offer a useful complement to traditional parametric SFH fitting and could inform quenching models.

major comments (2)
  1. [§3.3 and §4.1] §3.3 and §4.1: The validation of the score-based diffusion model (e.g., recovery of held-out photometry, posterior calibration on mocks, or explicit marginalisation over the IRAC Ch.1<26 selection function) is not described in sufficient detail. Since all quantitative results (SFRD, quiescent fractions, SFH correlations, AGN timing) rest on the accuracy of the inferred 16-dimensional SPS posteriors, this validation is load-bearing and must be strengthened to rule out selection biases or misspecification for low-mass/high-z galaxies.
  2. [§5.2, Figure 7] §5.2, Figure 7: The reported ~1 Gyr quenching timescale and mass-assembly concentration in the first 3.5 Gyr for 10<log M*/M⊙<11 galaxies are derived from the correlation matrices of the non-parametric SFHs. It is unclear how sensitive these timescales are to the choice of SFH basis or time binning; an explicit test with an alternative basis would be needed to confirm the result is not an artifact of the representation.
minor comments (2)
  1. [Abstract and §2] The abstract and §2 should explicitly state the number of diffusion steps, noise schedule, and any conditioning on redshift or magnitude used during training.
  2. [Table 1 or §4.3] Table 1 or §4.3: The comparison of quiescent fractions between sSFR and NUVrJ selections would benefit from a quantitative breakdown by stellar mass and redshift bin rather than the current summary statement.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their constructive and detailed report. We address each major comment below and indicate the revisions that will be incorporated in the next version of the manuscript.

read point-by-point responses

  1. Referee: [§3.3 and §4.1] §3.3 and §4.1: The validation of the score-based diffusion model (e.g., recovery of held-out photometry, posterior calibration on mocks, or explicit marginalisation over the IRAC Ch.1<26 selection function) is not described in sufficient detail. Since all quantitative results (SFRD, quiescent fractions, SFH correlations, AGN timing) rest on the accuracy of the inferred 16-dimensional SPS posteriors, this validation is load-bearing and must be strengthened to rule out selection biases or misspecification for low-mass/high-z galaxies.

    Authors: We agree that the validation of the generative model requires a more explicit and comprehensive presentation. In the revised manuscript we have substantially expanded §3.3 to include quantitative recovery statistics for held-out photometry across the full redshift and stellar-mass range, together with posterior calibration tests performed on mock catalogues that explicitly incorporate the IRAC Ch.1<26 selection function. A new subsection in §4.1 now reports bias and coverage diagnostics specifically for the low-mass and high-redshift regimes, demonstrating that any residual selection-induced biases remain well within the quoted uncertainties on the derived quantities. These additions directly strengthen the load-bearing validation. revision: yes

  2. Referee: [§5.2, Figure 7] §5.2, Figure 7: The reported ~1 Gyr quenching timescale and mass-assembly concentration in the first 3.5 Gyr for 10<log M*/M⊙<11 galaxies are derived from the correlation matrices of the non-parametric SFHs. It is unclear how sensitive these timescales are to the choice of SFH basis or time binning; an explicit test with an alternative basis would be needed to confirm the result is not an artifact of the representation.

    Authors: We acknowledge that robustness to the choice of non-parametric basis and time binning should be demonstrated explicitly. In the revised version we have added an alternative SFH representation that uses a different set of time bins (finer sampling in the most recent 2 Gyr). The corresponding correlation matrices are shown in a new panel of Figure 7 and discussed in §5.2. The ~1 Gyr quenching timescale and the concentration of mass assembly within the first 3.5 Gyr remain consistent to within 10 percent, confirming that the reported features are not artifacts of the original basis choice. revision: yes

Circularity Check

0 steps flagged

Generative diffusion model yields downstream SFRD and SFH statistics without definitional reduction to training inputs

full rationale

The paper trains a score-based diffusion model to match the joint distribution of 16 SPS parameters to the observed 26-band photometry of the IRAC-selected COSMOS2020 sample. SFRD is obtained by integrating the SFR values drawn from the learned generative distribution; SF/Q classifications and correlation matrices are computed directly on the inferred non-parametric SFHs. These quantities are downstream inferences from the fitted model rather than quantities defined by construction from the same parameters used in training. No load-bearing step reduces an output to an input via self-definition, fitted-parameter renaming, or a self-citation chain that lacks independent verification. Minor self-citation of prior diffusion-model methodology is present but not central to the reported SFRD peak or quenching timescales. The derivation remains self-contained against external photometric benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claims rest on the assumption that the diffusion model faithfully represents the underlying galaxy population and that the chosen sSFR threshold cleanly separates physical states; no new particles or forces are introduced.

axioms (2)
  • domain assumption The sSFR < 10^{-11} yr^{-1} threshold provides a physically meaningful separation between star-forming and quiescent galaxies that is less contaminated than NUVrJ color selection.
    Used to derive quiescent fractions and stellar mass functions that differ from color-based results.
  • domain assumption The 26-band photometry and IRAC magnitude limit allow the generative model to recover unbiased distributions over the 16 SPS parameters across the redshift range studied.
    Underlies the training and subsequent integration to obtain SFRD and SFHs.

pith-pipeline@v0.9.0 · 6008 in / 1728 out tokens · 52257 ms · 2026-05-18T13:59:07.820268+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Thermostats, Not Engines: A New Picture of Halo Gas Regulation

    astro-ph.CO 2026-05 unverdicted novelty 6.0

    Black hole feedback sets an entropy ceiling in halos enabling buoyant outflows with mass-independent entropy at the virial radius in simulations, predicting star formation rejuvenation above M_crit ~ 10^13.5-14 solar masses.

  2. The limits of feedback from active galactic nuclei

    astro-ph.GA 2026-05 unverdicted novelty 6.0

    AGN feedback creates a mass-independent entropy ceiling that allows outflows to escape halos only below M_200m = 10^13.7 M_sun, explaining depleted gas in groups versus near-cosmic fractions in clusters.

  3. KiDS+VIKING-450 cosmology with Bayesian hierarchical model redshift distributions

    astro-ph.CO 2026-05 conditional novelty 4.0

    Bayesian hierarchical modeling of photometric redshifts in KiDS+VIKING-450 raises S8 to 0.756 ± 0.039 and reduces Planck tension to 1.9σ.

Reference graph

Works this paper leans on

273 extracted references · 273 canonical work pages · cited by 3 Pith papers · 35 internal anchors

  1. [1]

    write newline

    " write newline "" before.all 'output.state := FUNCTION fin.entry write newline FUNCTION new.block output.state before.all = 'skip after.block 'output.state := if FUNCTION new.sentence output.state after.block = 'skip output.state before.all = 'skip after.sentence 'output.state := if if FUNCTION not #0 #1 if FUNCTION and 'skip pop #0 if FUNCTION or pop #1...

    work page

  2. [2]

    E., Gladders M

    Abramson L. E., Gladders M. D., Dressler A., Oemler Jr. A., Poggianti B., Vulcani B., 2015, @doi [ ] 10.1088/2041-8205/801/1/L12 , https://ui.adsabs.harvard.edu/abs/2015ApJ...801L..12A 801, L12

    work page doi:10.1088/2041-8205/801/1/l12 2015

  3. [3]

    Alberts S., Noble A., 2022, @doi [Universe] 10.3390/universe8110554 , https://ui.adsabs.harvard.edu/abs/2022Univ....8..554A 8, 554

    work page doi:10.3390/universe8110554 2022

  4. [4]

    Alsing J., et al., 2020, @doi [ ] 10.3847/1538-4365/ab917f , https://ui.adsabs.harvard.edu/abs/2020ApJS..249....5A 249, 5

    work page doi:10.3847/1538-4365/ab917f 2020

  5. [5]

    Alsing J., Peiris H., Mortlock D., Leja J., Leistedt B., 2023, @doi [ ] 10.3847/1538-4365/ac9583 , https://ui.adsabs.harvard.edu/abs/2023ApJS..264...29A 264, 29

    work page doi:10.3847/1538-4365/ac9583 2023

  6. [6]

    V., Leistedt B., Mortlock D., Leja J., 2024, @doi [ ] 10.3847/1538-4365/ad5c69 , https://ui.adsabs.harvard.edu/abs/2024ApJS..274...12A 274, 12

    Alsing J., Thorp S., Deger S., Peiris H. V., Leistedt B., Mortlock D., Leja J., 2024, @doi [ ] 10.3847/1538-4365/ad5c69 , https://ui.adsabs.harvard.edu/abs/2024ApJS..274...12A 274, 12

    work page doi:10.3847/1538-4365/ad5c69 2024

  7. [7]

    Arnouts S., Cristiani S., Moscardini L., Matarrese S., Lucchin F., Fontana A., Giallongo E., 1999, @doi [ ] 10.1046/j.1365-8711.1999.02978.x , https://ui.adsabs.harvard.edu/abs/1999MNRAS.310..540A 310, 540

    work page doi:10.1046/j.1365-8711.1999.02978.x 1999

  8. [8]

    Arnouts S., et al., 2013, @doi [ ] 10.1051/0004-6361/201321768 , https://ui.adsabs.harvard.edu/abs/2013A&A...558A..67A 558, A67

    work page doi:10.1051/0004-6361/201321768 2013

  9. [9]

    , keywords =

    Baldry I. K., Balogh M. L., Bower R. G., Glazebrook K., Nichol R. C., Bamford S. P., Budavari T., 2006, @doi [ ] 10.1111/j.1365-2966.2006.11081.x , https://ui.adsabs.harvard.edu/abs/2006MNRAS.373..469B 373, 469

    work page doi:10.1111/j.1365-2966.2006.11081.x 2006

  10. [10]

    L., Navarro J

    Balogh M. L., Navarro J. F., Morris S. L., 2000, @doi [ ] 10.1086/309323 , https://ui.adsabs.harvard.edu/abs/2000ApJ...540..113B 540, 113

    work page doi:10.1086/309323 2000

  11. [11]

    L., et al., 2016, @doi [ ] 10.1093/mnras/stv2949 , https://ui.adsabs.harvard.edu/abs/2016MNRAS.456.4364B 456, 4364

    Balogh M. L., et al., 2016, @doi [ ] 10.1093/mnras/stv2949 , https://ui.adsabs.harvard.edu/abs/2016MNRAS.456.4364B 456, 4364

    work page doi:10.1093/mnras/stv2949 2016

  12. [12]

    E., Hopkins, A

    Bauer A. E., et al., 2013, @doi [ ] 10.1093/mnras/stt1011 , https://ui.adsabs.harvard.edu/abs/2013MNRAS.434..209B 434, 209

    work page doi:10.1093/mnras/stt1011 2013

  13. [13]

    The Average Star Formation Histories of Galaxies in Dark Matter Halos from z=0-8

    Behroozi P. S., Wechsler R. H., Conroy C., 2013, @doi [ ] 10.1088/0004-637X/770/1/57 , https://ui.adsabs.harvard.edu/abs/2013ApJ...770...57B 770, 57

    work page internal anchor Pith review doi:10.1088/0004-637x/770/1/57 2013

  14. [14]

    H., Hearin, A

    Behroozi P., Wechsler R. H., Hearin A. P., Conroy C., 2019, @doi [ ] 10.1093/mnras/stz1182 , https://ui.adsabs.harvard.edu/abs/2019MNRAS.488.3143B 488, 3143

    work page doi:10.1093/mnras/stz1182 2019

  15. [15]

    J., Shioya Y., 2002, @doi [ ] 10.1086/342221 , https://ui.adsabs.harvard.edu/abs/2002ApJ...577..651B 577, 651

    Bekki K., Couch W. J., Shioya Y., 2002, @doi [ ] 10.1086/342221 , https://ui.adsabs.harvard.edu/abs/2002ApJ...577..651B 577, 651

    work page doi:10.1086/342221 2002

  16. [16]

    B., Ellis R

    Belli S., Newman A. B., Ellis R. S., 2019, @doi [ ] 10.3847/1538-4357/ab07af , https://ui.adsabs.harvard.edu/abs/2019ApJ...874...17B 874, 17

    work page doi:10.3847/1538-4357/ab07af 2019

  17. [17]

    Bellstedt S., et al., 2020, @doi [ ] 10.1093/mnras/staa2620 , https://ui.adsabs.harvard.edu/abs/2020MNRAS.498.5581B 498, 5581

    work page doi:10.1093/mnras/staa2620 2020

  18. [18]

    Bellstedt S., Robotham A. S. G., Driver S. P., Lagos C. d. P., Davies L. J. M., Cook R. H. W., 2024, @doi [ ] 10.1093/mnras/stae394 , https://ui.adsabs.harvard.edu/abs/2024MNRAS.528.5452B 528, 5452

    work page doi:10.1093/mnras/stae394 2024

  19. [19]

    G., Tal, T., et al

    Bezanson R., van Dokkum P. G., Tal T., Marchesini D., Kriek M., Franx M., Coppi P., 2009, @doi [ ] 10.1088/0004-637X/697/2/1290 , https://ui.adsabs.harvard.edu/abs/2009ApJ...697.1290B 697, 1290

    work page doi:10.1088/0004-637x/697/2/1290 2009

  20. [20]

    Cuadra and P

    Bosch Van Den van den Bosch F. C., Aquino D., Yang X., Mo H. J., Pasquali A., McIntosh D. H., Weinmann S. M., Kang X., 2008, @doi [ ] 10.1111/j.1365-2966.2008.13230.x , https://ui.adsabs.harvard.edu/abs/2008MNRAS.387...79V 387, 79

    work page doi:10.1111/j.1365-2966.2008.13230.x 2008

  21. [21]

    Boselli A., Fossati M., Sun M., 2022, @doi [ ] 10.1007/s00159-022-00140-3 , https://ui.adsabs.harvard.edu/abs/2022A&ARv..30....3B 30, 3

    work page doi:10.1007/s00159-022-00140-3 2022

  22. [22]

    J., et al., 2012, @doi [ ] 10.1088/0004-637X/754/2/83 , https://ui.adsabs.harvard.edu/abs/2012ApJ...754...83B 754, 83

    Bouwens R. J., et al., 2012, @doi [ ] 10.1088/0004-637X/754/2/83 , https://ui.adsabs.harvard.edu/abs/2012ApJ...754...83B 754, 83

    work page doi:10.1088/0004-637x/754/2/83 2012

  23. [23]

    J., Oesch, P

    Bouwens R. J., et al., 2021, @doi [ ] 10.3847/1538-3881/abf83e , https://ui.adsabs.harvard.edu/abs/2021AJ....162...47B 162, 47

    work page doi:10.3847/1538-3881/abf83e 2021

  24. [24]

    Bouwens R., Illingworth G., Oesch P., Stefanon M., Naidu R., van Leeuwen I., Magee D., 2023, @doi [ ] 10.1093/mnras/stad1014 , https://ui.adsabs.harvard.edu/abs/2023MNRAS.523.1009B 523, 1009

    work page doi:10.1093/mnras/stad1014 2023

  25. [25]

    G., Schaye, J., Frenk, C

    Bower R. G., Schaye J., Frenk C. S., Theuns T., Schaller M., Crain R. A., McAlpine S., 2017, @doi [ ] 10.1093/mnras/stw2735 , https://ui.adsabs.harvard.edu/abs/2017MNRAS.465...32B 465, 32

    work page doi:10.1093/mnras/stw2735 2017

  26. [26]

    Brand A., Allen L., Altman M., Hlava M., Scott J., 2015, @doi [Learned Publishing] 10.1087/20150211 , 28, 151

    work page doi:10.1087/20150211 2015

  27. [27]

    Buat V., et al., 2008, @doi [ ] 10.1051/0004-6361:20078263 , https://ui.adsabs.harvard.edu/abs/2008A&A...483..107B 483, 107

    work page doi:10.1051/0004-6361:20078263 2008

  28. [28]

    J., Conroy, C., & Johnson, B

    Byler N., Dalcanton J. J., Conroy C., Johnson B. D., 2017, @doi [ ] 10.3847/1538-4357/aa6c66 , https://ui.adsabs.harvard.edu/abs/2017ApJ...840...44B 840, 44

    work page doi:10.3847/1538-4357/aa6c66 2017

  29. [29]

    Calabr \`o A., et al., 2021, @doi [ ] 10.1051/0004-6361/202039244 , https://ui.adsabs.harvard.edu/abs/2021A&A...646A..39C 646, A39

    work page doi:10.1051/0004-6361/202039244 2021

  30. [30]

    C., et al

    Calzetti D., Armus L., Bohlin R. C., Kinney A. L., Koornneef J., Storchi-Bergmann T., 2000, @doi [ ] 10.1086/308692 , https://ui.adsabs.harvard.edu/abs/2000ApJ...533..682C 533, 682

    work page internal anchor Pith review doi:10.1086/308692 2000

  31. [31]

    Caplar N., Tacchella S., 2019, @doi [ ] 10.1093/mnras/stz1449 , https://ui.adsabs.harvard.edu/abs/2019MNRAS.487.3845C 487, 3845

    work page doi:10.1093/mnras/stz1449 2019

  32. [32]

    C., McLure, R

    Carnall A. C., McLure R. J., Dunlop J. S., Dav \'e R., 2018, @doi [ ] 10.1093/mnras/sty2169 , https://ui.adsabs.harvard.edu/abs/2018MNRAS.480.4379C 480, 4379

    work page internal anchor Pith review doi:10.1093/mnras/sty2169 2018

  33. [33]

    C., Leja J., Johnson B

    Carnall A. C., Leja J., Johnson B. D., McLure R. J., Dunlop J. S., Conroy C., 2019, @doi [ ] 10.3847/1538-4357/ab04a2 , https://ui.adsabs.harvard.edu/abs/2019ApJ...873...44C 873, 44

    work page doi:10.3847/1538-4357/ab04a2 2019

  34. [34]

    M., Hickox R

    Carroll C. M., Hickox R. C., Masini A., Lanz L., Assef R. J., Stern D., Chen C.-T. J., Ananna T. T., 2021, @doi [ ] 10.3847/1538-4357/abd185 , https://ui.adsabs.harvard.edu/abs/2021ApJ...908..185C 908, 185

    work page doi:10.3847/1538-4357/abd185 2021

  35. [35]

    Chabrier G., 2003, @doi [ ] 10.1086/376392 , https://ui.adsabs.harvard.edu/abs/2003PASP..115..763C 115, 763

    work page internal anchor Pith review doi:10.1086/376392 2003

  36. [36]

    Chaikin E., et al., 2025, preprint, https://ui.adsabs.harvard.edu/abs/2025arXiv250904067C ( @eprint arXiv 2509.04067 )

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  37. [37]

    A Simple Model for the Absorption of Starlight by Dust in Galaxies

    Charlot S., Fall S. M., 2000, @doi [ ] 10.1086/309250 , https://ui.adsabs.harvard.edu/abs/2000ApJ...539..718C 539, 718

    work page internal anchor Pith review doi:10.1086/309250 2000

  38. [38]

    B., Rudie G

    Chartab N., Newman A. B., Rudie G. C., Blanc G. A., Kelson D. D., 2024, @doi [ ] 10.3847/1538-4357/ad0554 , https://ui.adsabs.harvard.edu/abs/2024ApJ...960...73C 960, 73

    work page doi:10.3847/1538-4357/ad0554 2024

  39. [39]

    Chen R. T. Q., Rubanova Y., Bettencourt J., Duvenaud D. K., 2018, in Bengio S., Wallach H., Larochelle H., Grauman K., Cesa-Bianchi N., Garnett R., eds, Advances in Neural Information Processing Systems 31. pp 6572--6583 ( @eprint arXiv 1806.07366 )

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  40. [40]

    Ch \'e rouvrier D., et al., 2025, @doi [ ] 10.1051/0004-6361/202554450 , https://ui.adsabs.harvard.edu/abs/2025A&A...700A..30C 700, A30

    work page doi:10.1051/0004-6361/202554450 2025

  41. [41]

    Chevallard J., Charlot S., Wandelt B., Wild V., 2013, @doi [ ] 10.1093/mnras/stt523 , https://ui.adsabs.harvard.edu/abs/2013MNRAS.432.2061C 432, 2061

    work page doi:10.1093/mnras/stt523 2013

  42. [42]

    Chiang Y.-K., Makiya R., M \'e nard B., 2025, preprint, https://ui.adsabs.harvard.edu/abs/2025arXiv250405384C ( @eprint arXiv 2504.05384 )

    work page arXiv 2025

  43. [43]

    The Astrophysical Journal , year =

    Choi J., Dotter A., Conroy C., Cantiello M., Paxton B., Johnson B. D., 2016, @doi [ ] 10.3847/0004-637X/823/2/102 , https://ui.adsabs.harvard.edu/abs/2016ApJ...823..102C 823, 102

    work page internal anchor Pith review doi:10.3847/0004-637x/823/2/102 2016

  44. [44]

    A., Schnurr, O., Hirschi, R., et al

    Cid Fernandes R., Mateus A., Sodr \'e L., Stasi \'n ska G., Gomes J. M., 2005, @doi [ ] 10.1111/j.1365-2966.2005.08752.x , https://ui.adsabs.harvard.edu/abs/2005MNRAS.358..363C 358, 363

    work page doi:10.1111/j.1365-2966.2005.08752.x 2005

  45. [45]

    Ciesla L., Elbaz D., Fensch J., 2017, @doi [ ] 10.1051/0004-6361/201731036 , https://ui.adsabs.harvard.edu/abs/2017A&A...608A..41C 608, A41

    work page doi:10.1051/0004-6361/201731036 2017

  46. [46]

    Civano F., et al., 2016, @doi [ ] 10.3847/0004-637X/819/1/62 , https://ui.adsabs.harvard.edu/abs/2016ApJ...819...62C 819, 62

    work page doi:10.3847/0004-637x/819/1/62 2016

  47. [47]

    Conroy C., 2013, @doi [ ] 10.1146/annurev-astro-082812-141017 , https://ui.adsabs.harvard.edu/abs/2013ARA&A..51..393C 51, 393

    work page internal anchor Pith review doi:10.1146/annurev-astro-082812-141017 2013

  48. [48]

    E., 2010, @doi [ ] 10.1088/0004-637X/712/2/833 , https://ui.adsabs.harvard.edu/abs/2010ApJ...712..833C 712, 833

    Conroy C., Gunn J. E., 2010, @doi [ ] 10.1088/0004-637X/712/2/833 , https://ui.adsabs.harvard.edu/abs/2010ApJ...712..833C 712, 833

    work page internal anchor Pith review doi:10.1088/0004-637x/712/2/833 2010

  49. [49]

    E., & White , M

    Conroy C., Gunn J. E., White M., 2009, @doi [ ] 10.1088/0004-637X/699/1/486 , https://ui.adsabs.harvard.edu/abs/2009ApJ...699..486C 699, 486

    work page internal anchor Pith review doi:10.1088/0004-637x/699/1/486 2009

  50. [50]

    E., 2010, @doi [ ] 10.1088/0004-637X/708/1/58 , https://ui.adsabs.harvard.edu/abs/2010ApJ...708...58C 708, 58

    Conroy C., White M., Gunn J. E., 2010, @doi [ ] 10.1088/0004-637X/708/1/58 , https://ui.adsabs.harvard.edu/abs/2010ApJ...708...58C 708, 58

    work page doi:10.1088/0004-637x/708/1/58 2010

  51. [51]

    Cresci G., Mannucci F., Curti M., 2019, @doi [ ] 10.1051/0004-6361/201834637 , https://ui.adsabs.harvard.edu/abs/2019A&A...627A..42C 627, A42

    work page doi:10.1051/0004-6361/201834637 2019

  52. [52]

    Cucciati O., et al., 2012, @doi [ ] 10.1051/0004-6361/201118010 , https://ui.adsabs.harvard.edu/abs/2012A&A...539A..31C 539, A31

    work page doi:10.1051/0004-6361/201118010 2012

  53. [53]

    Cullen F., et al., 2019, @doi [ ] 10.1093/mnras/stz1402 , https://ui.adsabs.harvard.edu/abs/2019MNRAS.487.2038C 487, 2038

    work page doi:10.1093/mnras/stz1402 2019

  54. [54]

    Cunha Da da Cunha E., Charlot S., Elbaz D., 2008, @doi [ ] 10.1111/j.1365-2966.2008.13535.x , https://ui.adsabs.harvard.edu/abs/2008MNRAS.388.1595D 388, 1595

    work page doi:10.1111/j.1365-2966.2008.13535.x 2008

  55. [55]

    Curti M., Mannucci F., Cresci G., Maiolino R., 2020, @doi [ ] 10.1093/mnras/stz2910 , https://ui.adsabs.harvard.edu/abs/2020MNRAS.491..944C 491, 944

    work page doi:10.1093/mnras/stz2910 2020

  56. [56]

    Curti M., et al., 2024, @doi [ ] 10.1051/0004-6361/202346698 , https://ui.adsabs.harvard.edu/abs/2024A&A...684A..75C 684, A75

    work page doi:10.1051/0004-6361/202346698 2024

  57. [57]

    Curtis-Lake E., Chevallard J., Charlot S., Sandles L., 2021, @doi [ ] 10.1093/mnras/stab698 , https://ui.adsabs.harvard.edu/abs/2021MNRAS.503.4855C 503, 4855

    work page doi:10.1093/mnras/stab698 2021

  58. [58]

    DESI Collaboration et al., 2024, @doi [ ] 10.3847/1538-3881/ad3217 , https://ui.adsabs.harvard.edu/abs/2024AJ....168...58D 168, 58

    work page doi:10.3847/1538-3881/ad3217 2024

  59. [59]

    DESI Collaboration et al., 2025, preprint, https://ui.adsabs.harvard.edu/abs/2025arXiv250314745D ( @eprint arXiv 2503.14745 )

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  60. [60]

    D'Silva J. C. J., et al., 2023, @doi [ ] 10.1093/mnras/stad1974 , https://ui.adsabs.harvard.edu/abs/2023MNRAS.524.1448D 524, 1448

    work page doi:10.1093/mnras/stad1974 2023

  61. [61]

    Daddi E., Cimatti A., Renzini A., Fontana A., Mignoli M., Pozzetti L., Tozzi P., Zamorani G., 2004, @doi [ ] 10.1086/425569 , https://ui.adsabs.harvard.edu/abs/2004ApJ...617..746D 617, 746

    work page doi:10.1086/425569 2004

  62. [62]

    Daddi E., et al., 2007, @doi [ ] 10.1086/521818 , https://ui.adsabs.harvard.edu/abs/2007ApJ...670..156D 670, 156

    work page doi:10.1086/521818 2007

  63. [63]

    Davidzon I., et al., 2013, @doi [ ] 10.1051/0004-6361/201321511 , https://ui.adsabs.harvard.edu/abs/2013A&A...558A..23D 558, A23

    work page doi:10.1051/0004-6361/201321511 2013

  64. [64]

    Davidzon I., et al., 2017, @doi [ ] 10.1051/0004-6361/201730419 , https://ui.adsabs.harvard.edu/abs/2017A&A...605A..70D 605, A70

    work page doi:10.1051/0004-6361/201730419 2017

  65. [65]

    Elsevier, pp 282--311 ( @eprint arXiv 2502.01724 ), @doi 10.1016/B978-0-443-21439-4.00077-8

    De Lucia G., Fontanot F., Hirschmann M., Xie L., 2025, in Mandel I., McGee S., eds, Encyclopedia of Astrophysics Vol.\ 4. Elsevier, pp 282--311 ( @eprint arXiv 2502.01724 ), @doi 10.1016/B978-0-443-21439-4.00077-8

    work page doi:10.1016/b978-0-443-21439-4.00077-8 2025

  66. [66]

    Di Matteo T., Springel V., Hernquist L., 2005, @doi [ ] 10.1038/nature03335 , https://ui.adsabs.harvard.edu/abs/2005Natur.433..604D 433, 604

    work page internal anchor Pith review doi:10.1038/nature03335 2005

  67. [67]

    E., Torrey P., 2017, @doi [ ] 10.3847/1538-4357/aa68e5 , https://ui.adsabs.harvard.edu/abs/2017ApJ...839...26D 839, 26

    Diemer B., Sparre M., Abramson L. E., Torrey P., 2017, @doi [ ] 10.3847/1538-4357/aa68e5 , https://ui.adsabs.harvard.edu/abs/2017ApJ...839...26D 839, 26

    work page doi:10.3847/1538-4357/aa68e5 2017

  68. [68]

    Dom \' nguez S \'a nchez H., et al., 2011, @doi [ ] 10.1111/j.1365-2966.2011.19263.x , https://ui.adsabs.harvard.edu/abs/2011MNRAS.417..900D 417, 900

    work page doi:10.1111/j.1365-2966.2011.19263.x 2011

  69. [69]

    T., et al., 2023, @doi [ ] 10.1093/mnras/stac3472 , https://ui.adsabs.harvard.edu/abs/2023MNRAS.518.6011D 518, 6011

    Donnan C. T., et al., 2023, @doi [ ] 10.1093/mnras/stac3472 , https://ui.adsabs.harvard.edu/abs/2023MNRAS.518.6011D 518, 6011

    work page doi:10.1093/mnras/stac3472 2023

  70. [70]

    Dotter A., 2016, @doi [ ] 10.3847/0067-0049/222/1/8 , https://ui.adsabs.harvard.edu/abs/2016ApJS..222....8D 222, 8

    work page Pith review doi:10.3847/0067-0049/222/1/8 2016

  71. [71]

    Infrared Emission from Interstellar Dust. IV. The Silicate-Graphite-PAH Model in the Post-Spitzer Era

    Draine B. T., Li A., 2007, @doi [ ] 10.1086/511055 , https://ui.adsabs.harvard.edu/abs/2007ApJ...657..810D 657, 810

    work page Pith review doi:10.1086/511055 2007

  72. [72]

    B., et al., 2015, @doi [ ] 10.1093/mnras/stv2027 , https://ui.adsabs.harvard.edu/abs/2015MNRAS.454.2015D 454, 2015

    Drake A. B., et al., 2015, @doi [ ] 10.1093/mnras/stv2027 , https://ui.adsabs.harvard.edu/abs/2015MNRAS.454.2015D 454, 2015

    work page doi:10.1093/mnras/stv2027 2015

  73. [73]

    P., et al., 2018, @doi [ ] 10.1093/mnras/stx2728 , https://ui.adsabs.harvard.edu/abs/2018MNRAS.475.2891D 475, 2891

    Driver S. P., et al., 2018, @doi [ ] 10.1093/mnras/stx2728 , https://ui.adsabs.harvard.edu/abs/2018MNRAS.475.2891D 475, 2891

    work page doi:10.1093/mnras/stx2728 2018

  74. [74]

    P., et al

    Driver S. P., et al., 2022, @doi [ ] 10.1093/mnras/stac472 , https://ui.adsabs.harvard.edu/abs/2022MNRAS.513..439D 513, 439

    work page doi:10.1093/mnras/stac472 2022

  75. [75]

    Dunne L., et al., 2009, @doi [ ] 10.1111/j.1365-2966.2008.13900.x , https://ui.adsabs.harvard.edu/abs/2009MNRAS.394....3D 394, 3

    work page doi:10.1111/j.1365-2966.2008.13900.x 2009

  76. [76]

    L., Catinella B., Cortese L., 2018, @doi [ ] 10.1093/mnras/sty1247 , https://ui.adsabs.harvard.edu/abs/2018MNRAS.478.3447E 478, 3447

    Ellison S. L., Catinella B., Cortese L., 2018, @doi [ ] 10.1093/mnras/sty1247 , https://ui.adsabs.harvard.edu/abs/2018MNRAS.478.3447E 478, 3447

    work page doi:10.1093/mnras/sty1247 2018

  77. [77]

    R., 2024, @doi [Open J.\ Astrophys.] 10.33232/001c.127779 , https://ui.adsabs.harvard.edu/abs/2024OJAp....7E.121E 7, 121

    Ellison S., Ferreira L., Wild V., Wilkinson S., Rowlands K., Patton D. R., 2024, @doi [Open J.\ Astrophys.] 10.33232/001c.127779 , https://ui.adsabs.harvard.edu/abs/2024OJAp....7E.121E 7, 121

    work page doi:10.33232/001c.127779 2024

  78. [78]

    P., Topping M

    Endsley R., Chisholm J., Stark D. P., Topping M. W., Whitler L., 2025, @doi [ ] 10.3847/1538-4357/addc74 , https://ui.adsabs.harvard.edu/abs/2025ApJ...987..189E 987, 189

    work page doi:10.3847/1538-4357/addc74 2025

  79. [79]

    Euclid Collaboration et al., 2025, preprint, https://ui.adsabs.harvard.edu/abs/2025arXiv250417867E ( @eprint arXiv 2504.17867 )

    work page arXiv 2025

  80. [80]

    Observational Evidence of AGN Feedback

    Fabian A. C., 2012, @doi [ ] 10.1146/annurev-astro-081811-125521 , https://ui.adsabs.harvard.edu/abs/2012ARA&A..50..455F 50, 455

    work page internal anchor Pith review doi:10.1146/annurev-astro-081811-125521 2012

Showing first 80 references.