arxiv: 2605.11064 · v1 · submitted 2026-05-11 · 🌌 astro-ph.GA

Recognition: 2 theorem links

· Lean Theorem

Clustering constraints on super-early galaxy formation scenarios

Andrea Ferrara, Andrea Pallottini, Antonio Matteri

Authors on Pith no claims yet

Pith reviewed 2026-05-13 01:02 UTC · model grok-4.3

classification 🌌 astro-ph.GA

keywords galaxy clusteringhigh-redshift galaxiesJWST observationsUV luminosity functiongalaxy biasearly galaxy formationprimordial black holesstar formation models

0 comments

The pith

Galaxy clustering measurements can distinguish between models explaining the bright high-redshift galaxies seen by JWST.

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

The paper tests whether the way galaxies cluster in space at redshifts around 11 can reveal which physical process accounts for the unexpectedly bright early galaxies observed by JWST. Four scenarios that each match the observed UV luminosity function are applied to halos in a dark-matter simulation: one without attenuation, one with feedback-free bursts, one with bursty star formation, and one involving primordial black holes. The resulting galaxy bias is calculated for samples of different brightness, showing that faint galaxies cluster similarly across all models but brighter ones produce clearly different clustering strengths. This difference arises because each scenario links galaxy brightness to halo mass in its own way, so accurate future measurements could identify the correct explanation.

Core claim

Using the Shin-Uchuu dark-matter-only simulation to populate z ≈ 11 halos with galaxies according to attenuation-free, feedback-free burst, bursty star formation, and primordial black hole models, all of which reproduce the observed UV luminosity function, the two-point correlation function yields similar bias values of approximately 7 for faint galaxies at M_UV ≈ -16. The models diverge at M_UV ≲ -18, with the primordial black hole scenario producing nearly luminosity-independent bias while the others show increasing bias that reaches approximately 14 at M_UV ≈ -19. Current observations cannot yet exclude any scenario at high , but measurements of complete samples with M_UV < -18 would take

What carries the argument

The luminosity-dependent galaxy bias obtained from the two-point correlation function after assigning galaxies to halos via each scenario's distinct halo-mass to M_UV relation.

If this is right

All models agree on bias at faint luminosities but diverge at the bright end.
The primordial black hole scenario alone predicts nearly constant bias across luminosities.
The other three scenarios predict bias rising with luminosity up to about 14 at M_UV ≈ -19.
Future JWST data on complete samples brighter than M_UV = -18 can exploit these differences.
Refined theoretical predictions including baryonic effects will be needed to interpret the observations.

Where Pith is reading between the lines

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

This clustering approach could be applied to other high-redshift observables where multiple models fit the luminosity function but differ in spatial distribution.
Favoring one model would imply that feedback, bursts, or black holes shape not only galaxy numbers but also their large-scale placement at early times.
Combining bias measurements with other probes such as galaxy sizes might tighten constraints beyond what clustering alone provides.

Load-bearing premise

The chosen relations between halo mass and UV luminosity for each of the four scenarios correctly reflect the true physics, and the dark-matter simulation captures the clustering without needing baryonic corrections.

What would settle it

A precise measurement showing galaxy bias remaining near 7 at M_UV ≈ -19 with little increase toward brighter magnitudes would favor the primordial black hole model and weaken the others.

Figures

Figures reproduced from arXiv: 2605.11064 by Andrea Ferrara, Andrea Pallottini, Antonio Matteri.

**Figure 1.** Figure 1: Luminosity functions (LF) at z ≈ 11 predicted by the attenuation-free (AFM), feedback-free burst (FFB), bursty star formation (BSF), and primordial black hole (PBH) models (see Sec. 2.1 for details). Each model adopts the dark matter halo population from the Shin-Uchuu simulation Ishiyama et al. (2021) at z = 10.64, with the gray dotted line marking its volume limit (one galaxy per magnitude). The LF is … view at source ↗

**Figure 3.** Figure 3: The 2-point correlation function (ξ, 2PCF, eq. 18) of z ≈ 11 galaxies brighter than MUV = −19 for all models (see Sec. 2.1). For each model, ξ is fitted with a power-law to compute the corresponding bias (Sec. 2.4). For visualization sake, ξ is plotted multiplied by r 2 , so that a rising (descending) line means that the negative power-law index (γ, eq. 21) is smaller (larger) than 2. in the plot, the mor… view at source ↗

**Figure 4.** Figure 4: Bias estimate for all the adopted models (see Sec. [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗

read the original abstract

The unexpectedly high abundance of bright, blue, super-early galaxies ($z\gtrsim10$) has challenged most pre-JWST models of early galaxy formation and motivated a wide range of proposed explanations. We systematically investigate whether galaxy clustering can discriminate among representative scenarios that reproduce the observed UV luminosity function. Using the Shin-Uchuu dark-matter-only simulation, we populate $z \approx 11$ halos with galaxies according to solutions based on i) attenuation-free, ii) feedback-free bursts, iii) bursty star formation, and iv) primordial black hole models. For each model, we compute the two-point correlation function and predict the galaxy bias for flux-limited samples at different thresholds in the $-20 < {\rm M_{UV}} < -16$ magnitude range. We find that all models predict similar bias values ($b \approx 7$) for faint galaxies (${\rm M_{UV}}\approx-16$), but diverge at ${\rm M_{UV}}\lesssim-18$, as the underlying halo-mass to ${\rm M_{UV}}$ relations differ significantly. In particular, the primordial black hole scenario predicts an almost luminosity-independent bias, whereas the other models generally predict increasing bias with luminosity, reaching $b \approx 14$ for ${\rm M_{UV}} \approx -19$. Current observational estimates of the bias cannot yet rule out any of the models at a significant statistical confidence. More precise measurements from future JWST programs, together with improved theoretical predictions, will be required to break the present degeneracies. Ideally, constraints from a complete sample of galaxies with ${\rm M_{UV}} < -18$ would probe the knee of the $b({\rm M_{UV}})$ function, taking advantage of the difference in model predictions and strengthening our analysis. Although requiring further refinement, galaxy clustering is confirmed to be a promising probe of the physical origin of the JWST high-redshift luminosity function.

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.

Desk Editor's Note private letter to a colleague

The paper shows four JWST galaxy scenarios give similar faint-end bias but diverge at M_UV ≲ -18 when painted onto the same simulation, yet the split rests on untested halo mappings.

read the letter

The one thing to know is that these models predict different clustering strengths for the brightest galaxies at z~11, with the black hole one staying low while others get higher bias. The paper does a straightforward job of applying the same simulation to four different scenarios that were already tuned to the luminosity function. That gives clean, comparable bias numbers across them. The result that they agree at faint end but diverge at bright end is the concrete output. What is less solid is the reliance on the specific halo-mass to UV-magnitude relations taken from the source papers. No sensitivity tests are mentioned, so we don't know how much the divergence depends on exact parameter choices. The dark-matter-only run also leaves out possible baryonic impacts on clustering at high redshift, which could affect the bright end where rare halos matter most. This work is aimed at people following the JWST early galaxy debate who want to see what an independent observable like clustering might say. It is worth sending for peer review because the core comparison is done properly and the idea has merit, though the authors should address how robust the differences are to the input mappings.

Referee Report

2 major / 1 minor

Summary. The paper claims that populating the Shin-Uchuu dark-matter-only N-body simulation at z≈11 with galaxies according to four literature-derived M_UV-halo mass relations (attenuation-free, feedback-free bursts, bursty star formation, and primordial black holes) yields similar galaxy biases (b≈7) for faint samples (M_UV≈-16) but strong divergence at brighter luminosities (M_UV≲-18), with the PBH model nearly luminosity-independent while others reach b≈14; this suggests clustering measurements can discriminate the physical origin of the JWST high-z luminosity function, though current data cannot yet rule out models.

Significance. If the bias divergence is robust, the work establishes galaxy clustering as a useful complementary observable to the UV luminosity function for testing super-early galaxy formation scenarios. It supplies explicit, model-specific predictions for flux-limited samples across -20 < M_UV < -16 and identifies the bright-end regime where differences are largest, which can inform JWST observing strategies. The direct use of a large public simulation to compute the two-point correlation function provides a reproducible baseline for such tests.

major comments (2)

[Methods (halo population)] The halo-population procedure (described in the methods) imports distinct M_UV-halo mass relations for each of the four scenarios directly from the cited literature papers without any parameter variation, re-derivation, or propagation of their uncertainties into the correlation-function or bias results. Consequently the reported divergence at M_UV ≲ -18 is not demonstrated to be robust within this manuscript.
[Results (bias predictions)] All bias predictions rest on the dark-matter-only Shin-Uchuu simulation. At z≈11 the omission of baryonic processes (gas cooling, stellar feedback, possible AGN) can modify both the halo mass function and the spatial clustering of the halos that host UV-bright galaxies; any model-dependent baryonic suppression would alter the effective bias and could reduce or erase the claimed divergence between b≈14 and luminosity-independent values.

minor comments (1)

[Abstract] The abstract states the magnitude range as -20 < M_UV < -16 yet highlights divergence only for M_UV ≲ -18; a brief clarification of the exact luminosity thresholds used for the faint-end agreement (b≈7) would improve readability.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their thoughtful and constructive report. The comments have helped us identify areas where the manuscript can be clarified and strengthened, particularly regarding methodological limitations. We address each major comment below and indicate the revisions we will make.

read point-by-point responses

Referee: [Methods (halo population)] The halo-population procedure (described in the methods) imports distinct M_UV-halo mass relations for each of the four scenarios directly from the cited literature papers without any parameter variation, re-derivation, or propagation of their uncertainties into the correlation-function or bias results. Consequently the reported divergence at M_UV ≲ -18 is not demonstrated to be robust within this manuscript.

Authors: We appreciate the referee highlighting this methodological choice. The M_UV-halo mass relations were taken directly from the published literature to faithfully represent the distinct physical assumptions of each scenario (attenuation-free, feedback-free bursts, bursty star formation, and primordial black holes). Our focus was on comparing the clustering implications of these existing relations rather than re-fitting or varying parameters within them. We agree that a full propagation of uncertainties would provide a more quantitative assessment of robustness. In the revised manuscript, we will add explicit statements in the Methods and Discussion sections noting this limitation and clarifying that the reported divergence reflects the central relations from the literature. We will also suggest that future analyses could incorporate parameter variations to test sensitivity. revision: partial
Referee: [Results (bias predictions)] All bias predictions rest on the dark-matter-only Shin-Uchuu simulation. At z≈11 the omission of baryonic processes (gas cooling, stellar feedback, possible AGN) can modify both the halo mass function and the spatial clustering of the halos that host UV-bright galaxies; any model-dependent baryonic suppression would alter the effective bias and could reduce or erase the claimed divergence between b≈14 and luminosity-independent values.

Authors: This is a valid and important caveat. The Shin-Uchuu dark-matter-only simulation was selected for its large volume and resolution, which are necessary to reliably measure clustering for the rare bright galaxies at z≈11. The differences in bias between models arise from the distinct halo-mass selections implied by each M_UV-halo mass relation applied to the same underlying halo catalog. We acknowledge that baryonic processes could modify the halo mass function and clustering in a model-dependent way, potentially affecting the absolute bias values and the degree of divergence. In the revised manuscript, we will add a dedicated paragraph in the Discussion section addressing this limitation, explaining that our results provide a baseline using dark-matter clustering and that hydrodynamical simulations would be required to fully assess baryonic impacts. This does not change our conclusion that clustering can serve as a complementary probe, but it underscores the need for more advanced modeling. revision: yes

Circularity Check

0 steps flagged

No significant circularity; bias predictions are computed from external inputs.

full rationale

The paper takes the Shin-Uchuu dark-matter-only simulation as an external input and populates its z≈11 halos using M_UV–halo-mass relations taken directly from the four cited scenario papers. It then computes the two-point correlation function and bias values for flux-limited samples in the stated magnitude range. No equation or procedure inside the paper fits any parameter to clustering data, re-derives the halo-mass mappings, or defines bias in terms of itself; the reported divergence at M_UV ≲ −18 simply reflects the differences among the imported relations. The derivation chain is therefore self-contained against external benchmarks and contains no self-definitional, fitted-input, or self-citation-load-bearing steps.

Axiom & Free-Parameter Ledger

1 free parameters · 2 axioms · 0 invented entities

The central claim rests on the fidelity of the halo population step for each scenario and on the assumption that the dark-matter simulation provides an unbiased clustering template at z=11.

free parameters (1)

halo-mass to M_UV mapping per scenario
Each of the four models requires a relation between halo mass and UV magnitude chosen to reproduce the observed luminosity function; these mappings are not derived from first principles within the paper.

axioms (2)

domain assumption Shin-Uchuu N-body simulation accurately represents halo clustering at z≈11
The paper adopts the simulation output as the base for galaxy placement without additional baryonic corrections.
standard math Two-point correlation function of populated galaxies directly yields the linear bias
Standard cosmological assumption used to convert measured clustering to bias values.

pith-pipeline@v0.9.0 · 5654 in / 1659 out tokens · 55769 ms · 2026-05-13T01:02:44.857071+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

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

IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear
Using the Shin-Uchuu dark-matter-only simulation, we populate z≈11 halos with galaxies according to solutions based on i) attenuation-free, ii) feedback-free bursts, iii) bursty star formation, and iv) primordial black hole models. … compute the two-point correlation function and predict the galaxy bias
IndisputableMonolith/Foundation/DimensionForcing.lean alexander_duality_circle_linking unclear
the bias … b=σ_{8,g}/σ_8 … fitted with a power-law

Reference graph

Works this paper leans on

89 extracted references · 89 canonical work pages

[1]

J., Conselice, C

Adams, N. J., Conselice, C. J., Ferreira, L., et al. 2023, MNRAS, 518, 4755 Arrabal Haro, P., Dickinson, M., Bagley, M., et al. 2022, JWST Proposal. Cycle 1, 2750 Arrabal Haro, P., Dickinson, M., Finkelstein, S. L., et al. 2023, ApJL, 951, L22 Astropy Collaboration, Price-Whelan, A. M., Lim, P. L., et al. 2022, ApJL, 935, 167 Astropy Collaboration, Price-...

work page 2023
[2]

J., et al

Atek, H., Shuntov, M., Furtak, L. J., et al. 2023, MNRAS, 519, 1201

work page 2023
[3]

H., Hearin, A

Behroozi, P., Wechsler, R. H., Hearin, A. P., & Conroy, C. 2019, MNRAS, 488, 3143

work page 2019
[4]

J., Illingworth, G., Ellis, R

Bouwens, R. J., Illingworth, G., Ellis, R. S., Oesch, P., & Stefanon, M. 2022, ApJL, 940, 55

work page 2022
[5]

2025, MNRAS, 538, 3210

Boylan-Kolchin, M. 2025, MNRAS, 538, 3210

work page 2025
[6]

J., Saxena, A., Cameron, A

Bunker, A. J., Saxena, A., Cameron, A. J., et al. 2023, A&A, 677, A88

work page 2023
[7]

2024, Nature, 633, 318

Carniani, S., Hainline, K., D’Eugenio, F., et al. 2024, Nature, 633, 318

work page 2024
[8]

2025, A&A, 704, A290

Carvajal-Bohorquez, C., Ciesla, L., Laporte, N., et al. 2025, A&A, 704, A290

work page 2025
[9]

M., Akins, H

Casey, C. M., Akins, H. B., Shuntov, M., et al. 2024, ApJL, 965, 98

work page 2024
[10]

2025, A&A, 704, A158

Castellano, M., Fontana, A., Merlin, E., et al. 2025, A&A, 704, A158

work page 2025
[11]

2022, ApJL, 938, L15

Castellano, M., Fontana, A., Treu, T., et al. 2022, ApJL, 938, L15

work page 2022
[12]

2024, ApJL, 972, 143

Castellano, M., Napolitano, L., Fontana, A., et al. 2024, ApJL, 972, 143

work page 2024
[13]

& Choudhury, T

Chakraborty, A. & Choudhury, T. R. 2026, JCAP, 2026, 008

work page 2026
[14]

J., et al

Chemerynska, I., Atek, H., Furtak, L. J., et al. 2026, MNRAS, 546, staf2267

work page 2026
[15]

J., Adams, N., Harvey, T., et al

Conselice, C. J., Adams, N., Harvey, T., et al. 2025, ApJL, 983, 30

work page 2025
[16]

2023, Nature Astronomy, 7, 622

Curtis-Lake, E., Carniani, S., Cameron, A., et al. 2023, Nature Astronomy, 7, 622

work page 2023
[17]

M., & Trenti, M

Dalmasso, N., Ferrami, G., Leethochawalit, N., Ventura, E. M., & Trenti, M. 2026, MNRAS, 546, stag001

work page 2026
[18]

2024, MNRAS, 533, 2391

Dalmasso, N., Leethochawalit, N., Trenti, M., & Boyett, K. 2024, MNRAS, 533, 2391

work page 2024
[19]

S., & Pacucci, F

Dayal, P., Ferrara, A., Dunlop, J. S., & Pacucci, F. 2014, MNRAS, 445, 2545

work page 2014
[20]

C., Birnboim, Y ., Mandelker, N., & Li, Z

Dekel, A., Sarkar, K. C., Birnboim, Y ., Mandelker, N., & Li, Z. 2023, MNRAS, 523, 3201

work page 2023
[21]

2013, MNRAS, 435, 999

Dekel, A., Zolotov, A., Tweed, D., et al. 2013, MNRAS, 435, 999

work page 2013
[22]

2018, Physics Reports, 733, 1

Desjacques, V ., Jeong, D., & Schmidt, F. 2018, Physics Reports, 733, 1

work page 2018
[23]

T., McLure, R

Donnan, C. T., McLure, R. J., Dunlop, J. S., et al. 2024, MNRAS, 533, 3222

work page 2024
[24]

E., Villasenor, B., Robertson, B

Drakos, N. E., Villasenor, B., Robertson, B. E., et al. 2022, ApJL, 926, 194 Euclid Collaboration, Allen, N., Oesch, P. A., et al. 2025a Euclid Collaboration, Mellier, Y ., Abdurro’uf, et al. 2025b, A&A, 697, A1

work page 2022
[25]

2024, A&A, 684, A207

Ferrara, A. 2024, A&A, 684, A207

work page 2024
[26]

2023, MNRAS, 522, 3986

Ferrara, A., Pallottini, A., & Dayal, P. 2023, MNRAS, 522, 3986

work page 2023
[27]

L., Bagley, M

Finkelstein, S. L., Bagley, M. B., Arrabal Haro, P., et al. 2022, ApJL, 940, L55

work page 2022
[28]

L., Bagley, M

Finkelstein, S. L., Bagley, M. B., Ferguson, H. C., et al. 2023, ApJL, 946, L13

work page 2023
[29]

L., Leung, G

Finkelstein, S. L., Leung, G. C. K., Bagley, M. B., et al. 2024, ApJL, 969, L2

work page 2024
[30]

2023, ApJL, 949, L25

Fujimoto, S., Arrabal Haro, P., Dickinson, M., et al. 2023, ApJL, 949, L25

work page 2023
[31]

J., Shuntov, M., Atek, H., et al

Furtak, L. J., Shuntov, M., Atek, H., et al. 2023, MNRAS, 519, 3064

work page 2023
[32]

Gelli, V ., Mason, C., & Hayward, C. C. 2024, ApJL, 975, 192

work page 2024
[33]

N., Johnson, B

Hainline, K. N., Johnson, B. D., Robertson, B., et al. 2024, ApJL, 964, 71

work page 2024
[34]

2023, ApJS, 265, 5

Harikane, Y ., Ouchi, M., Oguri, M., et al. 2023, ApJS, 265, 5

work page 2023
[35]

R., Millman, K

Harris, C. R., Millman, K. J., van der Walt, S. J., et al. 2020, Nature, 585, 357

work page 2020
[36]

H., & Haghi, H

Haslbauer, M., Kroupa, P., Zonoozi, A. H., & Haghi, H. 2022, ApJL, 939, L31

work page 2022
[37]

Y .-Y ., Abdurro’uf, Coe, D., et al

Hsiao, T. Y .-Y ., Abdurro’uf, Coe, D., et al. 2024, ApJL, 973, 8

work page 2024
[38]

Hunter, J. D. 2007, Computing in Science and Engineering, 9, 90

work page 2007
[39]

R., Dayal, P., et al

Hutter, A., Cueto, E. R., Dayal, P., et al. 2025, A&A, 694, A254

work page 2025
[40]

A., et al

Ishiyama, T., Prada, F., Klypin, A. A., et al. 2021, MNRAS, 506, 4210

work page 2021
[41]

1984, ApJL, 284, L9

Kaiser, N. 1984, ApJL, 284, L9

work page 1984
[42]

2022, MNRAS, 511, 4005

Kannan, R., Garaldi, E., Smith, A., et al. 2022, MNRAS, 511, 4005

work page 2022
[43]

Kerscher, M., Szapudi, I., & Szalay, A. S. 2000, ApJL, 535, L13

work page 2000
[44]

& Cole, S

Lacey, C. & Cole, S. 1993, MNRAS, 262, 627

work page 1993
[45]

K., Pitrou, A., & Seibert, S

Lam, S. K., Pitrou, A., & Seibert, S. 2015, in Proc. Second Workshop on the LLVM Compiler Infrastructure in HPC, 1–6

work page 2015
[46]

Landy, S. D. & Szalay, A. S. 1993, ApJL, 412, 64

work page 1993
[47]

D., et al

Leitherer, C., Schaerer, D., Goldader, J. D., et al. 1999, ApJS, 123, 3

work page 1999
[48]

Leung, G. C. K., Bagley, M. B., Finkelstein, S. L., et al. 2023, ApJL, 954, L46

work page 2023
[49]

C., et al

Li, Z., Dekel, A., Sarkar, K. C., et al. 2024, A&A, 690, A108

work page 2024
[50]

& Bromm, V

Liu, B. & Bromm, V . 2022, ApJL, 937, L30

work page 2022
[51]

& Ferrara, A

Manzoni, D. & Ferrara, A. 2025, A&A, 704, L12

work page 2025
[52]

A., Trenti, M., & Treu, T

Mason, C. A., Trenti, M., & Treu, T. 2015, ApJL, 813, 21

work page 2015
[53]

A., Trenti, M., & Treu, T

Mason, C. A., Trenti, M., & Treu, T. 2023, MNRAS, 521, 497

work page 2023
[54]

McKerns, M. M. & Aivazis, M. A. G. 2010

work page 2010
[55]

M., Strand, L., Sullivan, T., Fang, A., & Aivazis, M

McKerns, M. M., Strand, L., Sullivan, T., Fang, A., & Aivazis, M. A. G. 2012

work page 2012
[56]

J., Donnan, C

McLeod, D. J., Donnan, C. T., McLure, R. J., et al. 2024, MNRAS, 527, 5004

work page 2024
[57]

2024, A&A, 691, A240

Merlin, E., Santini, P., Paris, D., et al. 2024, A&A, 691, A240

work page 2024
[58]

Mohammad, F. G. & Percival, W. J. 2022, MNRAS, 514, 1289 Muñoz, J. B., Chisholm, J., Sun, G., et al. 2026 Muñoz, J. B., Mirocha, J., Furlanetto, S., & Sabti, N. 2023, MNRAS, 526, L47

work page 2022
[59]

G., Power, C., & Robotham, A

Murray, S. G., Power, C., & Robotham, A. S. G. 2013, Astronomy and Comput- ing, 3, 23

work page 2013
[60]

P., Oesch, P

Naidu, R. P., Oesch, P. A., Brammer, G., et al. 2025

work page 2025
[61]

& Ferrara, A

Pallottini, A. & Ferrara, A. 2023, A&A, 677, L4

work page 2023
[62]

2025, A&A, 699, A6

Pallottini, A., Ferrara, A., Gallerani, S., et al. 2025, A&A, 699, A6

work page 2025
[63]

J., et al

Paquereau, L., Laigle, C., McCracken, H. J., et al. 2025, A&A, 702, A163

work page 2025
[64]

Peebles, P. J. E. 1980, The Large-Scale Structure of the Universe Pérez-González, P. G., Östlin, G., Costantin, L., et al. 2025, ApJL, 991, 179 Planck Collaboration, Ade, P. A. R., Aghanim, N., et al. 2016, A&A, 594, A13

work page 1980
[65]

D., Tacchella, S., et al

Robertson, B., Johnson, B. D., Tacchella, S., et al. 2024, ApJL, 970, 31

work page 2024
[66]

E., Tacchella, S., Johnson, B

Robertson, B. E., Tacchella, S., Johnson, B. D., et al. 2023, Nature Astronomy, 7, 611

work page 2023
[67]

2023, ApJL, 942, L27

Santini, P., Fontana, A., Castellano, M., et al. 2023, ApJL, 942, L27

work page 2023
[68]

F., Faucher-Giguère, C.-A., et al

Shen, X., Hopkins, P. F., Faucher-Giguère, C.-A., et al. 2020, MNRAS, 495, 3252

work page 2020
[69]

Shen, X., V ogelsberger, M., Boylan-Kolchin, M., Tacchella, S., & Naidu, R. P. 2024, MNRAS, 533, 3923

work page 2024
[70]

& Garrison, L

Sinha, M. & Garrison, L. H. 2019, in Software Challenges to Exascale Comput- ing (Singapore: Springer Singapore), 3–20

work page 2019
[71]

& Garrison, L

Sinha, M. & Garrison, L. H. 2020, MNRAS, 491, 3022

work page 2020
[72]

S., Yung, L

Somerville, R. S., Yung, L. Y . A., Lancaster, L., et al. 2025, MNRAS, 544, 3774

work page 2025
[73]

C., et al

Sun, G., Faucher-Giguère, C.-A., Hayward, C. C., et al. 2023, ApJL, 955, L35

work page 2023
[74]

D., Robertson, B

Tacchella, S., Johnson, B. D., Robertson, B. E., et al. 2023, MNRAS, 522, 6236

work page 2023
[75]

L., Robertson, B

Tinker, J. L., Robertson, B. E., Kravtsov, A. V ., et al. 2010, ApJL, 724, 878

work page 2010
[76]

2024, MNRAS, 529, 3563

Trinca, A., Schneider, R., Valiante, R., et al. 2024, MNRAS, 529, 3563

work page 2024
[77]

2024, MNRAS, 527, 9048

Trusov, S., Zarrouk, P., Cole, S., et al. 2024, MNRAS, 527, 9048

work page 2024
[78]

E., et al

Virtanen, P., Gommers, R., Oliphant, T. E., et al. 2020, Nature Medicine, 17, 261

work page 2020
[79]

2023, ApJL, 957, L34

Wang, B., Fujimoto, S., Labbé, I., et al. 2023, ApJL, 957, L34

work page 2023
[80]

Wechsler, R. H. & Tinker, J. L. 2018, ARA&A, 56, 435

work page 2018

Showing first 80 references.