Welcome!

Aside

Martin S. Sloth is a professor at the University of Southern Denmark (SDU) and group leader for the Universe Origins group. He is the first professor of theoretical cosmology in Denmark. Some of his selected research highlights are his proposal of the curvaton mechanism.  His work on non-Gaussianity and the discovery of the consistency relations for exchange diagrams. His work on IR effects in inflation and the discovery of the consistency relations for IR loop effects in de Sitter and inflationary spacetimes. Inflationary magnetogenesis and the consistency relations for cosmic magnetic fields.  Universal constraints on axions from inflation. The paradigm of Planckian Interacting Dark Matter (PIDM).  And most recently, the proposed solution to the Hubble tension, New Early Dark Energy (NEDE).

We mainly work on understanding the origin of the universe, dark matter, and dark energy. You can find more information about our research here. If you are interested in more popular descriptions of our research, you can look for some of our outreach activities here. Finally, on the group page, you can find a description of the group members.

E-mail: sloth@sdu.dk

Publications of M. S. Sloth: HEP-INSPIRE, Google Scholar.

M. S. Sloth on Linkedin

Posted on by

At a recent DESY (Hamburg) colloquium, I gave a review talk on “Resolving the Hubble Tension with New Early Dark Energy.”

I explained how the DESI anomaly might be connected to the Hubble tension and have an early-universe explanation in terms of dark acoustic oscillations. 

Slides + full video: https://physikseminar.desy.de/hamburg/colloquia_in_2025/16_december_2025/

The paper mentioned in the video clip below (now online): arXiv:2512.15870 — https://arxiv.org/abs/2512.15870

This builds on earlier work on New Early Dark Energy, with excellent collaborators: Florian Niedermann, Mathias Garny, Henrique Rubira, Aleksandr Chatrchyan, Juan Cruz, Emil Brinch Holm, Vivian Poulin, Thomas Tram, Steen Hannestad …

Posted on by

A new early-universe explanation of the DESI anomaly connecting to the predictions of our solution to the Hubble tension in collaboration with Mathias Garny and Florian Niedermann: https://arxiv.org/abs/2512.15870

The presence of a Dark Acoustic Oscillation (DAO) near the actual BAO may appear as a single BAO-only feature with a slightly shifted peak position, which could bias the DESI BAO measurements, thereby explaining the anomaly in the DESI DR2 data without invoking late-time-evolving Dark Energy. The DAO predicted in our model for solving the Hubble tension, Hot New Early Dark Energy (Hot NEDE) with Dark Radiation Matter Decoupling (DRMD), also meets the requirements for explaining the DESI results, as shown in the leftmost panel of the figure below. The green curve is the favoured DAO position by the DESI DR2 data on small scales, and the dotted black line is the prediction of DRMD in Hot NEDE, when solving the Hubble tension.

Screenshot 2025-12-19 at 07.52.54.png

Posted on by

New community report today uses different methods and techniques to establish the value of the Hubble constant to 1% precision, H0=73.50+/-0.81 km/s/Mpc, confirming the Hubble tension at more than 5-sigma.

This means that our current standard model of the Universe, the ΛCDM model, can not describe the real Universe as we observe it, and we must move to finding the correct model of our Universe replacing ΛCDM.

New Early Dark Energy (NEDE) has proven itself to be a promising framework in which to search for the solution to the Hubble tension (see previous posts), and we will keep intensifying our efforts to explore NEDE.

See the community report here: https://arxiv.org/abs/2510.23823

Posted on by

To make sense of the Hubble tension, it is essential to have a simple (in terms of ideas), convincing solution. In a new paper, we present a new microscopic Hot NEDE model of the dark sector based on well-known fundamental principles, gauge symmetry, and spontaneous symmetry breaking, which resolves the Hubble tension. We also discuss how this solution can be further tested with future precision data.

Profile likelihood curves for the model from our paper can be seen below.  

Work in collaboration with Mathias Garny, Florian Niedermann and Henrique Rubira.

The paper is on arXiv: https://arxiv.org/abs/2508.03795.

Posted on by

“Simplicity is part of what I mean by beauty, but it is a simplicity of ideas, not simplicity of a mechanical sort that can be measured by counting equations or symbols.”

— Steven Weinberg

Posted on by

In a new paper, “I demonstrate that soft graviton modes in de Sitter are the Goldstone modes of the spontaneously broken asymptotic symmetry group of de Sitter. I then show that any local measurement, including the effects of the environment, will collapse the symmetric state onto the broken state in the large volume limit. In any discussion involving observers, de Sitter is, therefore, best described globally by the broken phase, while local observers, in the small volume limit, can not discriminate between different degenerate global vacuum states, and are therefore best described by the symmetric state. This illuminates the physical nature of soft graviton modes in de Sitter.”

As a bonus, I also speculate about the possibility that a new type of crystals, with the same symmetry-breaking pattern as discussed in the paper, could be realized in the laboratory as an analog system of de Sitter and eternal inflation.

The paper is on arXiv:  https://arxiv.org/abs/2502.03520

Posted on by

In an update of our recent paper with Aleksandr Chatrchyan, Florian Niedermann, and Vivian Poulin, we have included fits of Cold New Early Dark Energy (Cold NEDE) to the DESI BAO data. 

Using  CMB Planck Plik data, DESI BAO, and Pantheon+ data, we find around 2-sigma evidence for NEDE (see upper left panel below), which is enhanced to 6-sigma evidence when including SH0ES supernovae data (see upper right panel below). 

Without including SH0ES data, a value of H0=73 km/s/Mpc, compatible with SH0ES findings, is allowed within 2-sigma using Planck Plik, DESI BAO, and Pantheon+ data (see lower left panel below). Including SH0ES data, the Lambda-CDM preferred value of H0 less than 68 km/s/Mpc is excluded at 6-sigma in the Cold NEDE model.

For likelihoods with other dataset combinations, parameters, and discussion, see our full paper: https://arxiv.org/abs/2408.14537.

Posted on by

In a new paper “we explore the possibility that exotic forms of dark matter could expose humans on Earth or on prolonged space travel to a significant radiation dose. The radiation exposure from dark matter interacting with nuclei in the human body is generally assumed to be negligible compared to other sources of background radiation. However, as we discuss here, current data allow for dark matter models where this is not necessarily true”. See: https://arxiv.org/abs/2411.10521

Posted on by

In a new paper with Aleksandr Chatrchyan, Florian Niedermann, and Vivian Poulin (https://arxiv.org/abs/2408.14537), we generalize the treatment of the NEDE fluid in the Cold NEDE model and test it against updated CMB and Supernovae data. We show that in the Cold NEDE model, the NEDE fluid can, as previously anticipated, be described as a mix of stiff fluid and radiation consistently with data. 

The new CMB Planck NPIPE data and Pantheon+ supernova data slightly increase the tension between ΛCDM and SH0ES. This similarly slightly increases the residual tension in (N)EDE models when relieving the Hubble tension, with Cold NEDE doing slightly better than AxiEDE and other EDE-type models. On the other hand, we know from other studies that the recent DESI data pulls in the opposite direction toward reducing the residual tension by a similar amount. In conclusion, further improvements in data and NEDE model building provide a constructive and promising way forward. 

Finally, we have made a non-trivial test of the Cold NEDE model. In the Cold NEDE model, it is a theoretical prediction that the ratio between the Hubble rate at the time of the phase transition, H*, and the mass of the trigger field, m, should satisfy H*/m ~ 0.2. Being agnostic about the underlying model and fitting H* and m as free parameters to the data is a way of testing this specific prediction of Cold NEDE. It is spectacular that we find a peak in the probability distribution for H*/m between 0.20 and 0.22 as predicted (see figure below).

Screenshot 2024-08-28 at 09.01.28.png