Recognition: unknown
Double the axions, half the tension: multi-field early dark energy eases the Hubble tension
Pith reviewed 2026-05-10 12:53 UTC · model grok-4.3
The pith
Two axion-like early dark energy fields relax Planck constraints and cut the Hubble tension to 1.5 sigma.
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
Core claim
We show that the strong constraints placed by Planck NPIPE Cosmic Microwave Background (CMB) data on axion-like early dark energy (EDE) are significantly alleviated in models with multiple fields. We find a 1.5 sigma residual tension with the Local Distance Network value of H0 in a 2-field model, with no improvement beyond two fields, and a best-fit value of H0 ~1.4 sigma larger than in the 1-field case. The second field improves the fit to high-ℓ CMB data, where 1-field EDE is most strongly disfavored, and suggests modifications to the pre-recombination history over a wider redshift range.
What carries the argument
Multi-field axion-like early dark energy, in which extra scalar fields inject energy before recombination over an extended redshift interval and thereby loosen CMB constraints on the Hubble constant.
If this is right
- Two-field models fit high-ℓ CMB data better than single-field models while raising the preferred H0.
- Residual tension with local measurements falls to 1.5 sigma.
- A third field produces no additional improvement.
- The required energy injection spans a broader redshift window before recombination.
Where Pith is reading between the lines
- The result implies that the Hubble tension may be sensitive to the number of light degrees of freedom active in the early universe.
- Future CMB polarization measurements at small scales could distinguish the two-field scenario from single-field EDE.
- The model must still be checked for consistency with baryon acoustic oscillation and supernova data not emphasized in the current analysis.
Load-bearing premise
The extra parameters supplied by the second field deliver real physical improvement rather than simply fitting noise or unaccounted systematics in the Planck NPIPE data.
What would settle it
A reanalysis of the Planck NPIPE high-ℓ data that removes the fit improvement once known systematics are modeled, or a new local H0 measurement that falls below the two-field best-fit value, would falsify the claimed alleviation.
Figures
read the original abstract
We show that the strong constraints placed by Planck NPIPE Cosmic Microwave Background (CMB) data on axion-like early dark energy (EDE) are significantly alleviated in models with multiple fields. We find a $1.5\sigma$ residual tension with the Local Distance Network value of $H_0$ in a 2-field model, with no improvement beyond two fields, and a best-fit value of $H_0$ $\sim 1.4\sigma$ larger than in the 1-field case. The second field improves the fit to high-$\ell$ CMB data, where 1-field EDE is most strongly disfavored, and suggests modifications to the pre-recombination history over a wider redshift range.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The manuscript presents a multi-field extension of axion-like early dark energy (EDE) models to address the Hubble tension. It shows that Planck NPIPE CMB data impose strong constraints on single-field EDE, but these are significantly relaxed in a two-field model, which improves the fit particularly to high-ℓ data. This yields a best-fit H0 ~1.4σ higher than the one-field case and reduces the residual tension with local distance ladder measurements to 1.5σ. No further improvement occurs with more than two fields, and the second field enables modifications to the pre-recombination expansion history over a wider redshift range.
Significance. If the results hold, the work is significant because it identifies a concrete mechanism—multiple axion-like fields—by which EDE can accommodate higher H0 values while remaining consistent with CMB observations that disfavor the single-field case. The finding that two fields are optimal and that the gain is concentrated at high-ℓ provides a clear target for theoretical model-building in axion cosmology. The direct numerical comparison of one- versus multi-field fits to the same NPIPE dataset is a strength, as is the emphasis on the redshift range of the EDE contribution.
major comments (2)
- [Abstract and §4] Abstract and §4 (results): the reported 1.4σ increase in best-fit H0 and reduction to 1.5σ tension are obtained from CMB-only fits; the manuscript does not show that the posterior remains consistent or that the tension reduction persists when BAO or Pantheon+ data are included, which is load-bearing because EDE models are known to be sensitive to these late-time anchors.
- [§3 and Table 2] §3 (methodology) and Table 2: each additional field introduces at least two new parameters (e.g., f_EDE,i and z_c,i or equivalent masses/initial values); the Δχ² improvement at high-ℓ is not accompanied by a reported Bayesian evidence ratio or information criterion that demonstrates the gain exceeds the Occam penalty, leaving open whether the alleviation is physical or an artifact of increased flexibility.
minor comments (2)
- [Figure 3] Figure 3: the legend and axis labels for the high-ℓ residual plots should explicitly state the multipole range used for the χ² contribution to make the claimed improvement at high-ℓ quantitatively traceable.
- [§2] Notation: the definition of the effective number of relativistic degrees of freedom or the sound horizon shift induced by each field should be given explicitly in §2 to avoid ambiguity when comparing one- and two-field cases.
Simulated Author's Rebuttal
We thank the referee for their careful reading of the manuscript and for the constructive comments. We address each major comment below.
read point-by-point responses
-
Referee: [Abstract and §4] Abstract and §4 (results): the reported 1.4σ increase in best-fit H0 and reduction to 1.5σ tension are obtained from CMB-only fits; the manuscript does not show that the posterior remains consistent or that the tension reduction persists when BAO or Pantheon+ data are included, which is load-bearing because EDE models are known to be sensitive to these late-time anchors.
Authors: We agree that the primary results and quoted H0 shift are from CMB-only fits to NPIPE data. The manuscript's focus is on demonstrating how the two-field extension relaxes the tight CMB constraints that disfavor single-field EDE, particularly at high-ℓ, thereby permitting a higher best-fit H0. While we recognize that late-time anchors can further constrain EDE, the wider redshift range enabled by the second field is intended to provide additional flexibility that may preserve consistency. In the revised manuscript we will add combined CMB+BAO fits (and a brief discussion of Pantheon+) to explicitly test whether the tension reduction persists. revision: yes
-
Referee: [§3 and Table 2] §3 (methodology) and Table 2: each additional field introduces at least two new parameters (e.g., f_EDE,i and z_c,i or equivalent masses/initial values); the Δχ² improvement at high-ℓ is not accompanied by a reported Bayesian evidence ratio or information criterion that demonstrates the gain exceeds the Occam penalty, leaving open whether the alleviation is physical or an artifact of increased flexibility.
Authors: The referee is correct that each additional field adds parameters and that we report only Δχ² improvements. We note, however, that the three-field model yields no further improvement over the two-field case, which already suggests the gain is not driven purely by extra degrees of freedom. To address the Occam penalty directly, the revised manuscript will include AIC values (AIC = χ² + 2k) for the one-, two-, and three-field models using the same data and parameter counts. A full nested-sampling evidence ratio remains computationally prohibitive at present but the AIC comparison will quantify whether the improvement justifies the added complexity. revision: yes
Circularity Check
No significant circularity: results from numerical fits to external CMB data
full rationale
The paper reports MCMC fits of multi-field axion-like EDE models to Planck NPIPE CMB data, showing improved high-ℓ fits and a higher best-fit H0 that reduces tension with the independent local distance ladder measurement. No load-bearing derivation chain reduces to self-definition, fitted inputs renamed as predictions, or self-citation chains; the central result is a statistical outcome of additional parameters on external data, not a closed tautology. The analysis remains falsifiable against BAO, supernovae, or other datasets not used in the primary fit.
Axiom & Free-Parameter Ledger
free parameters (1)
- masses and initial field values for each axion-like field
axioms (1)
- domain assumption Background and perturbation evolution follows standard FLRW cosmology plus scalar field dynamics
invented entities (1)
-
second axion-like scalar field
no independent evidence
Forward citations
Cited by 3 Pith papers
-
Geometric Constraints on the Pre-Recombination Expansion History from the Hubble Tension
Model-independent reconstruction shows that early-universe modifications resolving the Hubble tension exist at the background level, requiring a smooth ~15% pre-recombination expansion rate enhancement.
-
A barotropic alternative to Early Dark Energy for alleviating the $H_0$ tension
A barotropic fluid with ω_s ≈ 0.29 and Ω_s ≈ 1.5×10^{-5} raises the inferred H0 to match SH0ES while remaining consistent with Planck CMB, DESI BAO, and Pantheon data.
-
Breaking Free from the Swampland of Impossible Universes through the DESI Portal
DESI data indicating evolving dark energy may allow string theory to describe observed universes without violating swampland constraints on constant dark energy.
Reference graph
Works this paper leans on
-
[1]
Tensions between the Early and the Late Universe
L. Verde, T. Treu, and A. G. Riess, Nature Astron.3, 891 (2019), arXiv:1907.10625 [astro-ph.CO]
work page internal anchor Pith review arXiv 2019
-
[2]
E. Di Valentinoet al., Astropart. Phys.131, 102605 (2021), arXiv:2008.11284 [astro-ph.CO]
-
[3]
E. Di Valentino, O. Mena, S. Pan, L. Visinelli, W. Yang, A. Melchiorri, D. F. Mota, A. G. Riess, and J. Silk, Class. Quant. Grav.38, 153001 (2021), arXiv:2103.01183 [astro-ph.CO]
-
[4]
L. Perivolaropoulos and F. Skara, New Astron. Rev.95, 101659 (2022), arXiv:2105.05208 [astro-ph.CO]
- [5]
-
[6]
2022, JHEAp, 34, 49, doi: 10.1016/j.jheap.2022.04.002
E. Abdallaet al., JHEAp34, 49 (2022), arXiv:2203.06142 [astro-ph.CO]
work page internal anchor Pith review arXiv 2022
-
[7]
Di Valentino, Universe8, 399 (2022)
E. Di Valentino, Universe8, 399 (2022)
2022
-
[8]
J.-P. Hu and F.-Y. Wang, Universe9, 94 (2023), arXiv:2302.05709 [astro-ph.CO]
- [9]
-
[10]
E. Di Valentinoet al.(CosmoVerse Network), Phys. Dark Univ.49, 101965 (2025), arXiv:2504.01669 [astro- ph.CO]
work page internal anchor Pith review arXiv 2025
-
[11]
S. Casertanoet al.(H0DN), (2025), arXiv:2510.23823 [astro-ph.CO]
work page internal anchor Pith review Pith/arXiv arXiv 2025
-
[12]
E. Camphuiset al.(SPT-3G), (2025), arXiv:2506.20707 [astro-ph.CO]
work page internal anchor Pith review arXiv 2025
- [13]
-
[14]
N. Sch¨ oneberg, J. Lesgourgues, and D. C. Hooper, JCAP10, 029 (2019), arXiv:1907.11594 [astro-ph.CO]
-
[15]
DESI DR2 Results II: Measurements of Baryon Acoustic Oscillations and Cosmological Constraints
M. Abdul Karimet al.(DESI), Phys. Rev. D112, 083515 (2025), arXiv:2503.14738 [astro-ph.CO]
work page Pith review arXiv 2025
-
[16]
Efstathiou, (2020), arXiv:2007.10716 [astro-ph.CO]
G. Efstathiou, (2020), arXiv:2007.10716 [astro-ph.CO]
-
[17]
E. Mortsell, A. Goobar, J. Johansson, and S. Dhawan, Astrophys. J.935, 58 (2022), arXiv:2106.09400 [astro- ph.CO]
- [18]
-
[19]
D. Camarena, V. Marra, Z. Sakr, and C. Clark- son, Class. Quant. Grav.39, 184001 (2022), arXiv:2205.05422 [astro-ph.CO]
-
[20]
R. Wojtak and J. Hjorth, Mon. Not. Roy. Astron. Soc. 515, 2790 (2022), arXiv:2206.08160 [astro-ph.CO]
- [21]
- [22]
-
[23]
R. Wojtak and J. Hjorth, Mon. Not. Roy. Astron. Soc. 533, 2319 (2024), arXiv:2403.10388 [astro-ph.CO]
- [24]
-
[25]
L. Perivolaropoulos, Phys. Rev. D110, 123518 (2024), arXiv:2408.11031 [astro-ph.CO]
- [26]
-
[27]
Physically-motivated priors in the local distance ladder significantly reduce the Hubble tension
M. H¨ og˚ as and E. M¨ ortsell, (2026), arXiv:2601.22215 [astro-ph.CO]
work page internal anchor Pith review Pith/arXiv arXiv 2026
-
[28]
E. M¨ ortsell and S. Dhawan, JCAP09, 025 (2018), arXiv:1801.07260 [astro-ph.CO]
-
[29]
S. Vagnozzi, S. Dhawan, M. Gerbino, K. Freese, A. Goo- bar, and O. Mena, Phys. Rev. D98, 083501 (2018), arXiv:1801.08553 [astro-ph.CO]
- [30]
- [31]
- [32]
-
[33]
S. Vagnozzi, Phys. Rev. D102, 023518 (2020), arXiv:1907.07569 [astro-ph.CO]
-
[34]
L. Visinelli, S. Vagnozzi, and U. Danielsson, Symmetry 11, 1035 (2019), arXiv:1907.07953 [astro-ph.CO]
-
[35]
E. Di Valentino, A. Melchiorri, O. Mena, and S. Vagnozzi, Phys. Dark Univ.30, 100666 (2020), arXiv:1908.04281 [astro-ph.CO]
-
[36]
E. Di Valentino, A. Melchiorri, O. Mena, and S. Vagnozzi, Phys. Rev. D101, 063502 (2020), arXiv:1910.09853 [astro-ph.CO]
-
[37]
C. Krishnan, E. ´O. Colg´ ain, Ruchika, A. A. Sen, M. M. Sheikh-Jabbari, and T. Yang, Phys. Rev. D102, 103525 (2020), arXiv:2002.06044 [astro-ph.CO]
-
[38]
G. Alestas, L. Kazantzidis, and L. Perivolaropoulos, Phys. Rev. D101, 123516 (2020), arXiv:2004.08363 [astro-ph.CO]
-
[39]
K. Jedamzik and L. Pogosian, Phys. Rev. Lett.125, 181302 (2020), arXiv:2004.09487 [astro-ph.CO]
-
[40]
T. Sekiguchi and T. Takahashi, Phys. Rev. D103, 083507 (2021), arXiv:2007.03381 [astro-ph.CO]
-
[41]
S. Roy Choudhury, S. Hannestad, and T. Tram, JCAP 03, 084 (2021), arXiv:2012.07519 [astro-ph.CO]
-
[42]
T. Brinckmann, J. H. Chang, and M. LoVerde, Phys. Rev. D104, 063523 (2021), arXiv:2012.11830 [astro- ph.CO]
- [43]
-
[44]
V. Marra and L. Perivolaropoulos, Phys. Rev. D104, L021303 (2021), arXiv:2102.06012 [astro-ph.CO]
- [45]
-
[46]
C. Krishnan, R. Mohayaee, E. ´O. Colg´ ain, M. M. Sheikh-Jabbari, and L. Yin, Class. Quant. Grav.38, 184001 (2021), arXiv:2105.09790 [astro-ph.CO]
-
[47]
F.-Y. Cyr-Racine, F. Ge, and L. Knox, Phys. Rev. Lett. 128, 201301 (2022), arXiv:2107.13000 [astro-ph.CO]
- [48]
- [49]
- [50]
- [51]
-
[52]
N. Sch¨ oneberg and G. Franco Abell´ an, JCAP12, 001 (2022), arXiv:2206.11276 [astro-ph.CO]
-
[53]
S. Banerjee, M. Petronikolou, and E. N. Saridakis, Phys. Rev. D108, 024012 (2023), arXiv:2209.02426 [gr- qc]
-
[54]
R. de S´ a, M. Benetti, and L. L. Graef, Eur. Phys. J. Plus137, 1129 (2022), arXiv:2209.11476 [astro-ph.CO]
- [55]
- [56]
-
[57]
M. Khodadi and M. Schreck, Phys. Dark Univ.39, 101170 (2023), arXiv:2301.03883 [gr-qc]
- [58]
-
[59]
I. Ben-Dayan and U. Kumar, JCAP12, 047 (2023), arXiv:2302.00067 [astro-ph.CO]
-
[60]
A. G´ omez-Valent, N. E. Mavromatos, and J. Sol` a Per- acaula, Class. Quant. Grav.41, 015026 (2024), arXiv:2305.15774 [gr-qc]
-
[61]
Ruchika, H. Rathore, S. Roy Choudhury, and V. Rentala, JCAP06, 056 (2024), arXiv:2306.05450 [astro-ph.CO]
- [62]
-
[63]
E. Frion, D. Camarena, L. Giani, T. Miranda, D. Bertacca, V. Marra, and O. F. Piattella, (2023), 10.21105/astro.2307.06320, arXiv:2307.06320 [astro-ph.CO]
-
[64]
A. G´ omez-Valent, A. Favale, M. Migliaccio, and A. A. Sen, Phys. Rev. D109, 023525 (2024), arXiv:2309.07795 [astro-ph.CO]
- [65]
-
[66]
W. Giar` e, Y. Zhai, S. Pan, E. Di Valentino, R. C. Nunes, and C. van de Bruck, Phys. Rev. D110, 063527 (2024), arXiv:2404.02110 [astro-ph.CO]
- [67]
-
[68]
W. Giar` e, M. A. Sabogal, R. C. Nunes, and E. Di Valentino, Phys. Rev. Lett.133, 251003 (2024), arXiv:2404.15232 [astro-ph.CO]
- [69]
- [70]
- [71]
- [72]
-
[73]
S. Roy Choudhury and T. Okumura, Astrophys. J. Lett. 976, L11 (2024), arXiv:2409.13022 [astro-ph.CO]
- [74]
- [75]
- [76]
- [77]
- [78]
-
[79]
Y.-M. Zhang, T.-N. Li, G.-H. Du, S.-H. Zhou, L.- Y. Gao, J.-F. Zhang, and X. Zhang, (2025), arXiv:2510.12627 [astro-ph.CO]
-
[80]
S. Kumar, Phys. Dark Univ.52, 102248 (2026), arXiv:2512.19000 [astro-ph.CO]
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.