Tachyonic Encore: A universal shift of inflationary observables
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: High Energy Physics: Theory. The day in high-energy theory: gravity, strings, fields and the questions left open.
Kai: We're looking at the paper: "Tachyonic Encore: A universal shift of inflationary observables".
Mira: Here's the gist of the paper. A generic, largely inflaton-potential–independent mechanism in which a light axion spectator reshapes inflationary observables through purely gravitational multi-field dynamics is proposed to reconcile otherwise disfavored inflaton potentials with current CMB constraints Mechanism Overview The core idea involves a light axion spectator field initialized near the hilltop of its potential reshaping inflationary observables through purely gravitational multi-field dynamics During inflation the axion is frozen and the background follows an effectively single-field trajectory.
Kai: After inflation ends, the axion begins to roll, inducing a turn in field space and transient tachyonic phases of the isocurvature mode.
Mira: These resulting “tachyonic encore” occur entirely on super-horizon scales.
Kai: These phases generate a nearly scale-invariant enhancement of the curvature power spectrum, suppressing the tensor-to-scalar ratio and shifting the scalar tilt to a weighted combination of adiabatic and entropic tilts at horizon crossing.
Paper summary: Kai: So this paper, "Tachyonic Encore: A universal shift of inflationary observables," proposes this idea that a light axion spectator field can reshape how we see inflation through purely gravitational multi-field dynamics. It suggests that during inflation, this axion is essentially frozen near the hilltop of its potential and the background follows a single-field path, but after inflation ends, it starts to roll.
Mira: And when it rolls, it induces a turn in field space and these transient tachyonic phases in the isocurvature mode cause something called a "tachyonic encore." This happens entirely on super-horizon scales. The authors claim this process generates a nearly scale-invariant enhancement of the curvature power spectrum.
Lev: From what I see, the key is that these isocurvature modes can source curvature perturbations, which isn't usually how we think about it in single-field slow roll scenarios >
Kai: Exactly. The paper sets up this multi-field system using a quadratic Lagrangian for the perturbations of n scalar fields, and they define directions called tangential and orthogonal directions with respect to the background trajectory >
Mira: They simplify the field equations in the super-horizon limit by defining variables like R for curvature perturbation and S for isocurvature perturbation, which lets them see how these modes evolve even if the background trajectory turns >
Lev: I'm curious about what they need from this axion spectator field to make this work. What are the specific conditions they require?
Kai: The essential requirements are actually quite minimal. They say you just need effective single-field dynamics during inflation, meaning the background path is nearly straight, and then you need an axion spectator field frozen in a region of negative potential curvature >
Mira: A simple example they use is a potential like V theta equals lambda to the fourth over one minus cosine theta f, and the hierarchy ensures that this axion stays frozen while inflation is happening >
Lev: And then it has to start rolling shortly after inflation ends, with its mass comparable to the Hubble parameter at that time, and roll before reheating becomes efficient >
Kai: That timing is important because it means all this enhancement happens entirely on super-horizon scales, not during reheating >
Conclusion: Kai: Looking at the paper "Tachyonic Encore: A universal shift of inflationary observables," the main point is how this light axion spectator field can dramatically impact the power spectrum, tilt, and even non-Gaussianity of scalar fluctuations >
Mira: It’s about using purely gravitational multi-field dynamics to reconcile otherwise disfavored inflaton potentials with what we see in the CMB data >
Lev: The implication for us is that if this mechanism works, we should expect a specific signature: order-one local non-Gaussianity, which they predict to be within reach of upcoming surveys >
Kai: That nonlinear signature comes from the fact that the enhancement and its nonlinear behavior stem from the same post-inflationary axion dynamics, not some extra conversion process like a decay channel >
Mira: They show how the mass and a specific ratio in their potential determine how much that power spectrum shifts, which allows them to move towards larger values consistent with recent CMB data >
Lev: The paper also points out that neglecting reheating effects is justified in standard perturbative scenarios because the relevant scale is set by the spectator axion dynamics, not some very high reheating temperature like ten twelve GeV > <ref:2606.12549#pg1>
Kai: So, to put it plainly, this mechanism offers a way to adjust inflationary observables without needing a specific inflaton potential that might be hard to justify on its own >
Mira: It suggests that even simple setups can be tuned by these spectator fields to fit the observed cosmological parameters better >
Diogo S. Gorgulho, *Margherita Putti, ‡Rodrigo Gonzalez Quaglia, ‡Ema Dimastrogiovanni, Matteo Fasiello, Diederik Roest
Van Swinderen Institute for Particle Physics and Gravity University of Groningen Instituto de Ciencias Físicas Universidad Nacional Autónoma de Mîxico Instituto de Física T⠙rica UAM-CSIC
hep-th, astro-ph.CO, gr-qc, hep-ph
Submitted: 2026-06-10
Updated: 2026-11-04
Comments: 13 pages, 4 figures
License: http://creativecommons.org/licenses/by/4.0/
Importance score: 80/100
The gist: A generic, largely inflaton-potential–independent mechanism in which a light axion spectator reshapes inflationary observables through purely gravitational multi-field dynamics is proposed to…
Key concepts
- Axion Spectator
- A light axion field that is initialized near a potential hilltop during inflation. It remains frozen during inflation, acting as a spectator, but begins to roll after inflation ends. This rolling induces dynamics that affect the observable inflationary parameters.
- Tachyonic Encore
- Transient phases of tachyonic instability in the isocurvature mode that occur on super-horizon scales after inflation. These phases are caused by the axion spectator field beginning to roll, and they generate a nearly scale-invariant enhancement of the curvature power spectrum.
- Multi-field Dynamics
- The mathematical description of a system involving multiple scalar fields during inflation. The paper uses an orthonormal basis to separate these dynamics into adiabatic (tangential) and entropic (orthogonal) directions, showing how isocurvature modes can source curvature perturbations.
- Enhancement Parameter (E)
- A parameter that relates the power spectrum amplitude at the end of inflation ($A_0R$) to the amplitude after the enhancement ($AsR$). This parameter dictates how much the final scalar power spectrum is boosted, leading to a smaller tensor-to-scalar ratio compared to single-field models.
Terminology
Summary
A generic, largely inflaton-potential–independent mechanism in which a light axion spectator reshapes inflationary observables through purely gravitational multi-field dynamics is proposed to reconcile otherwise disfavored inflaton potentials with current CMB constraints
Mechanism Overview
The core idea involves a light axion spectator field initialized near the hilltop of its potential reshaping inflationary observables through purely gravitational multi-field dynamics During inflation the axion is frozen and the background follows an effectively single-field trajectory. After inflation ends, the axion begins to roll, inducing a turn in field space and transient tachyonic phases of the isocurvature mode. p. 0 During inflation the axion is frozen and the background
follows an effectively single-field trajectory. After inflation ends, the axion rolls, inducing a turn in
field space and transient tachyonic phases of the isocurvature mode. After inflation ends, the axion rolls, inducing a turn in
field space and transient tachyonic phases of the isocurvature mode. The resulting “tachyonic encore”
occurs entirely on super-horizon scales. The resulting “tachyonic encore”
occurs entirely on super-horizon scales. These phases generate a nearly scale-invariant enhancement
of the curvature power spectrum, suppressing the tensor-to-scalar ratio and shifting the scalar tilt
to a weighted combination of adiabatic and entropic tilts at horizon crossing.Where the paper says it
Where the paper says it
Where the paper says it
Multi-field Dynamics
The multi-field system is described by a quadratic Lagrangian for the perturbations QI of n scalar fields ϕI (in spatially-flat gauge). The mass components are defined as M2 IJ ≡ V;IJ − RIJKLϕ˙Kϕ˙L − 1/a3 Dt/a3 Hϕ˙Iϕ˙J. p. 1 2
with GIJ the field space metric and MPl = 1. The mass
components are defined as
M2
IJ ≡ V;IJ − RIJKLϕ˙Kϕ˙L −
1
a
3 Dt
a
3
H
ϕ˙
Iϕ˙
J
, (2)
with RIJKL the Riemann tensor of the field space. In order to best illustrate the analogy with the singlefield case, we define the tangential (adiabatic) and orthogonal (entropic) directions σ and s with respect to
the background trajectory [24]. Specialising to the twofield case, this orthonormal basis is given by
e
I
σ ≡
ϕ˙I
σ˙
, and e
I
s ≡ −G
−1/2
ϵ
IJ ϕ˙
J
σ˙
, (3)
where ˙σ
2 ≡ GIJ ϕ˙Iϕ˙J
, and adiabatic and entropic fluctuations are given by Qi = e
I
i QI with i = (σ, s). (6)
While physical observables are determined by the canonically normalized variable Qs, we will use S as proxy to
describe the evolution of perturbations. The field equations for curvature and isocurvature perturbations simplify in the super-horizon limit (k ≪ aH):
R˙ = 2
Ω
H
S +
C(k)
a
3ϵ
,
S¨ + (3+η)HS˙ + m2
S,effS = O
k
2
a
2
,
(7)
where C(k) is a k-dependent integration constant, whose
contribution rapidly redshifts and becomes negligible,
and the effective isocurvature mass is
m2
S,eff ≡ V;ss − V;σσ + ϵH2R +
1
a
3
d
dt
a
3σ˙
2
H
+ 4Ω2
, (8)
where R is the Ricci scalar of the field space metric. The field equations for curvature and isocurvature perturbations simplify in the super-horizon limit (k ≪ aH):
R˙ = 2
Ω
H
S +
C(k)
a
3ϵ
,
S¨ + (3+η)HS˙ + m2
S,effS = O
k
2
a
2
,
(7)
where C(k) is a k-dependent integration constant, whose
contribution rapidly redshifts and becomes negligible,
and the effective isocurvature mass is
m2
S,eff ≡ V;ss − V;σσ + ϵH2R +
1
a
3
d
dt
a
3σ˙
2
H
+ 4Ω2
, (8)
where R is the Ricci scalar of the field space metric. In
contrast to the single-field case, the curvature perturbation need not freeze on super-horizon scales: isocurvature modes can source curvature perturbations, leading
to non-trivial super-horizon evolution, provided the background trajectory turns (Ω ̸= 0).Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Axion Spectator Requirements
The essential requirements for the encore mechanism are remarkably minimal: effective single-field dynamics during inflation (omega ≃ 0) with an axion spectator field θ frozen in a region of negative potential curvature. A simple example is the axion potential V (θ) = Λ4 / (1 - cos θ f). p. 1 The essential requirements for
the encore mechanism are remarkably minimal: effective
single-field dynamics during inflation (Ω ≃ 0) with an
axion spectator field θ frozen in a region of negative potential curvature. A simple example is the axion potential
V (θ) = Λ4
1 − cos
θ
f
, (9)
where f is the axion decay constant. Our mechanism is
in place for an axion mass m2
θ = Vθθ < 0 satisfying1
Hreh ≲ |mθ| ≃ Hend ≪ Hhc , (10)
where Hreh, Hend and Hhc denote respectively the Hubble scale during reheating, at the end of inflation and
at horizon crossing of the CMB modes. This hierarchy ensures that the axion remains frozen during inflation (mθ ≪ Hhc), begins to roll shortly after its end
(mθ ≃ Hend), and does so before reheating becomes efficient (mθ ≳ Hreh), thereby allowing the enhancement
to occur entirely on super-horizon scales.Where the paper says it
Where the paper says it
Observational Consequences
The post-enhancement power spectrum is related to its value at the end of inflation via the enhancement parameter E ≡ As / A0R, where A0R and AsR are the amplitude of the scalar power spectrum at the end of inflation and after the enhancement, respectively. This leads to a smaller tensor-to-scalar ratio with respect to the single-field case rpost = PTs/PR = PT/E P0R as the gravitational sector is unaffected. p. 2 1 × 10−9
). This
leads to a smaller tensor-to-scalar ratio with respect to
the single-field case,
r
post =
PT
Ps
R
=
PT
E P0
R
, (18)
as the gravitational sector is unaffected. (24)
3 This contribution is essentially additive with respect to any other
non-Gaussianity stemming from inflationary (self-) interactions. It is well-know that derivative interactions support equilateral
and orthogonal type non-Gaussianity whilst standard single-field
slow-roll dynamics driven by a simple V (φ) leads to a slow roll
suppressed local bispectrum [44]. The
same dynamics predict local non-Gaussianity, f
loc. NL ∼ O(1), within reach of upcoming surveys.Where the paper says it
Where the paper says it
Where the paper says it
Modified Observables and Constraints
The final curvature perturbation inherits the spectral tilt of its two components in Eq. (14) evaluated at horizon crossing npost s = 1/E nR0 + E - 1/E nQs hc. p. 3 The final curvature perturbation therefore inherits the
spectral tilt of its two components in Eq. (14) evaluated
at horizon crossing2
n
post
s =
1
E
nR0 +
E − 1
E
nQs
hc
. (14) evaluated
at horizon crossing2
n
post
s =
1
E
nR0 +
E − 1
E
nQs
hc
. (19)
For E close to 1, this reproduces the usual single-field
inflation result with
nR0 − 1 = 2ησσ − 6ϵ , (20)
where ηαα ≡ V;αα/V . (19)
For E close to 1, this reproduces the usual single-field
inflation result with
nR0 − 1 = 2ησσ − 6ϵ , (20)
where ηαα ≡ V;αα/V . For E ≫ 1, the sourced component
dominates and the final tilt is dictated mostly by the
entropic mode. Given that the enhancement dynamics takes place at
super-horizon scales, we anticipate (and also verify numerically) that this mechanism engenders3 a local component for the bispectrum which is well-captured by the
δN formalism. Upon using the second order expansion
for R, one arrives at the bispectrum and obtains, for the
non-linear parameter4 fNL,
f
loc. NL =
5
6
"
1
E
2
Nφφ
N2
φ
+
2(E − 1)
E
2
Nφθ
NφNθ
+
E − 1
E
2 Nθθ
N2
θ
#
. (23)
In the E ≫ 1 regime, this expression simplifies to
f
loc. Notice that the encore is effective whenever
E ≫ 1. In this regime, n
post
s
is shifted from the standard single-field results by corrections controlled by m2
θ
/m2
φ (for
chaotic) and Λ4/V0 (for Starobinsky), r
post is suppressed, and f
loc.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Axiverse Context
The mechanism can be found within the string axiverse where the axion at the heart of the encore is part of a set of additional spectator sectors. The conditions for extra spectator axions (a ≥ 2) are relatively mild for typical values of the axion decay constant fa and energy scale Λa. p. 7 The
presence of several such axions makes the required hilltop initial condition and mass range substantially less
restrictive than in a one-spectator model. The conditions the extra spectator axions (a ≥ 2)
ought to satisfy in order not to impact the CMB are relatively mild for typical values of the axion decay constant
fa and energy scale Λa. This is because, despite the fact that
they can carry significant energy, these spectators start
from a generic phase and so lose the logarithmic sensitivity to ∆θ we have in the encore, which is key to the
enhancement, as encoded in Eq. The configurations θT fLΛL and θT fSΛL are the most
interesting. (36)
n
post
s ≃ 1 −
Λ
4
f
2V0
2 +
3
N
−
3
2N2
,
together with the non-Gaussian parameter in Eq. The mass and (Λ4/V0) ratio controls the shift in ns, making it possible to move towards larger values, ∆θ determines r while the COBE normalization sets the inflaton
mass and the potential scale correspondingly. (36)
n
post
s ≃ 1 −
Λ
4
f
2V0
2 +
3
N
−
3
2N2
,
together with the non-Gaussian parameter in Eq. The mass and (Λ4/V0) ratio controls the shift in ns, making it possible to move towards larger values, ∆θ determines r while the COBE normalization sets the inflaton
mass and the potential scale correspondingly. (36)
n
post
s ≃ 1 −
Λ
4
f
2V0
2 +
3
N
−
3
2N2
,
together with the non-Gaussian parameter in Eq. The mass and (Λ4/V0) ratio controls the shift in ns, making it possible to move towards larger values, ∆θ determines r while the COBE normalization sets the inflaton
mass and the potential scale correspondingly. (36)
n
post
s ≃ 1 −
Λ
4
f
2V0
2 +
3
N
−
3
2N2
,
together with the non-Gaussian parameter in Eq. The mass and (Λ4/V0) ratio controls the shift in ns, making it possible to move towards larger values, ∆θ determines r while the COBE normalization sets the inflaton
mass and the potential scale correspondingly. (36)
n
post
s ≃ 1 −
Λ
4
f
2V0
2 +
3
N
−
3
2N2
,
together with the non-Gaussian parameter in Eq. The mass and (Λ4/V0) ratio controls the shift in ns, making it possible to move towards larger values, ∆θ determines r while the COBE normalization sets the inflaton
mass and the potential scale correspondingly. (36)
n
post
s ≃ 1 −
Λ
4
f
2V0
2 +
3
N
−
3
2N2
,
together with the non-Gaussian parameter in Eq. The mass and (Λ4/V0) ratio controls the shift in ns, making it possible to move towards larger values, ∆θ determines r while the COBE normalization sets the inflaton
mass and the potential scale correspondingly. (36)
n
post
s ≃ 1 −
Λ
4
f
2V0
2 +
3
N
−
3
2N2
,
together with the non-Gaussian parameter in Eq. The mass and (Λ4/V0) ratio controls the shift in ns, making it possible to move towards larger values, ∆θ determines r while the COBE normalization sets the inflaton
mass and the potential scale correspondingly. (36)
n
post
s ≃ 1 −
Λ
4
f
2V0
2 +
3
N
−
3
2N2
,
together with the non-Gaussian parameter in Eq. The mass and (Λ4/V0) ratio controls the shift in ns, making it possible to move towards larger values, ∆θ determines r while the COBE normalization sets the inflaton
mass and the potential scale correspondingly. (36)
n
post
s ≃ 1 −
Λ
4
f
2V0
2 +
3
N
−
3
2N2
,
together with the non-Gaussian parameter in Eq. The mass and (Λ4/V0) ratio controls the shift in ns, making it possible to move towards larger values, ∆θ determines r while the COBE normalization sets the inflaton
mass and the potential scale correspondingly. (36)
n
post
s ≃ 1 −
Λ
4
f
2V0
2 +
3
N
−
3
2N2
,
together with the non-Gaussian parameter in Eq. The mass and (Λ4/V0) ratio controls the shift in ns, making it possible to move towards larger values, ∆θ determines r while the COBE normalization sets the inflaton
mass and the potential scale correspondingly. (36)
n
post
s ≃ 1 −
Λ
4
f
2V0
2 +
3
N
−
3
2N2
,
together with the non-Gaussian parameter in Eq. The mass and (Λ4/V0) ratio controls the shift in ns, making it possible to move towards larger values, ∆θ determines r while the COBE normalization sets the inflaton
mass and the potential scale correspondingly. (36)
n
post
s ≃ 1 −
Λ
4
f
2V0
2 +
3
N
−
3
2N2
,
together with the non-Gaussian parameter in Eq. The mass and (Λ4/V0) ratio controls the shift in ns, making it possible to move towards larger values, ∆θ determines r while the COBE normalization sets the inflaton
mass and the potential scale correspondingly. (36)
n
post
s ≃ 1 −
Λ
4
f
2V0
2 +
3
N
−
3
2N2
,
together with the non-Gaussian parameter in Eq. The mass and (Λ4/V0) ratio controls the shift in ns, making it possible to move towards larger values, ∆θ determines r while the COBE normalization sets the inflaton
mass and the potential scale correspondingly. (36)
n
post
s ≃ 1 −
Λ
4
f
2V0
2 +
3
N
−
3
2N2
,
together with the non-Gaussian parameter in Eq. The mass and (Λ4/V0) ratio controls the shift in ns, making it possible to move towards larger values, ∆θ determines r while the COBE normalization sets the inflaton
mass and the potential scale correspondingly. (36)
n
post
s ≃ 1 −
Λ
4
f
2V0
2 +
3
N
−
3
2N2
,
together with the non-Gaussian parameter in Eq. The mass and (Λ4/V0) ratio controls the shift in ns, making it possible to move towards larger values, ∆θ determines r while the COBE normalization sets the inflaton
mass and the potential scale correspondingly. (36)
n
post
s ≃ 1 −
Λ
4
f
2V0
2 +
3
N
−
3
2N2
,
together with the non-Gaussian parameter in Eq. The mass and (Λ4/V0) ratio controls the shift in ns, making it possible to move towards larger values, ∆θ determines r while the COBE normalization sets the inflaton
mass and the potential scale correspondingly. (36)
n
post
s ≃ 1 −
Λ
4
f
2V0
2 +
3
N
−
3
2N2
,
together with the non-Gaussian parameter in Eq. The mass and (Λ4/V0) ratio controls the shift in ns, making it possible to move towards larger values, ∆θ determines r while the COBE normalization sets the inflaton
mass and the potential scale correspondingly. (36)
n
post
s ≃ 1 −
Λ
4
f
2V0
2 +
3
N
−
3
2N2
,
together with the non-Gaussian parameter in Eq. The mass and (Λ4/V0) ratio controls the shift in ns, making it possible to move towards larger values, ∆θ determines r while the COBE normalization sets the inflaton
mass and the potential scale correspondingly. (36)
n
post
s ≃ 1 −
Λ
4
f
2V0
2 +
3
N
−
3
2N2
,
together with the non-Gaussian parameter in Eq. The mass and (Λ4/V0) ratio controls the shift in ns, making it possible to move towards larger values, ∆θ determines r while the COBE normalization sets the inflaton
mass and the potential scale correspondingly. (36)
n
post
s ≃ 1 −
Λ
4
f
2V0
2 +
3
N
−
3
2N2
,
together with the non-Gaussian parameter in Eq. The mass and (Λ4/V0) ratio controls the shift in ns, making it possible to move towards larger values, ∆θ determines r while the COBE normalization sets the inflaton
mass and the potential scale correspondingly. (36)
n
post
s ≃ 1 −
Λ
4
f
2V0
2 +
3
N
−
3
2N2
,
together with the non-Gaussian parameter in Eq. The mass and (Λ4/V0) ratio controls the shift in ns, making it possible to move towards larger values, ∆θ determines r while the COBE normalization sets the inflaton
mass and the potential scale correspondingly. (36)
n
post
s ≃ 1 −
Λ
4
f
2V0
2 +
3
N
−
3
2N2
,
together with the non-Gaussian parameter in Eq. The mass and (Λ4/V0) ratio controls the shift in ns, making it possible to move towards larger values, ∆θ determines r while the COBE normalization sets the inflaton
mass and the potential scale correspondingly. (36)
n
post
s ≃ 1 −
Λ
4
f
2V0
2 +
3
N
−
3
2N2
,
together with the non-Gaussian parameter in Eq. The mass and (Λ4/V0) ratio controls the shift in ns, making it possible to move towards larger values, ∆θ determines r while the COBE normalization sets the inflaton
mass and the potential scale correspondingly.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Reheating Dynamics
The analysis shows that neglecting reheating effects is justified within standard perturbative reheating scenarios where the decay rate Γ is subdominant to the Hubble parameter H(Npost). The relevant scale is set by the spectator axion dynamics with mθ ∼ Hend so the same result can equivalently be written in terms of the axion mass. p. 8 We thus obtain
the temperature at which reheating becomes dynamically
relevant
T
dyn
re ≃
90
π
2g∗
1/4
p
Hend e
− 3
4
(1+weff )Npost
. (33)
In the present setup, the relevant scale is set by the spectator axion dynamics with mθ ∼ Hend so the same result
can equivalently be written in terms of the axion mass. This
condition still leaves a large viable window for reheating
processes. Indeed, successful Big Bang nucleosynthesis
only requires Tre ≳ O(10) MeV, while for Npost = 10
one finds T
dyn
re ∼ 1012 GeV. Indeed, successful Big Bang nucleosynthesis
only requires Tre ≳ O(10) MeV, while for Npost = 10
one finds T
dyn
re ∼ 1012 GeV. Therefore, reheating temperatures several orders of magnitude below T
dyn
re remain comfortably compatible with the lower bound from
standard cosmology.Where the paper says it
Where the paper says it
Where the paper says it
Conclusion
In this Letter, a mechanism based on a spectator axion initialized near the hilltop of its potential can dramatically impact the power spectrum, tilt, and non-Gaussianity of scalar fluctuations. If overlaid on standard single-field slow-roll dynamics, our simple setup can reconcile well-known inflationary models with current experimental constraints. p. 4 In this Letter, we have presented a mechanism, based on a spectator axion initialized near the hilltop of its potential, whose post-inflationary dynamics can
dramatically impact the power spectrum, tilt, and nonGaussianity of scalar fluctuations. If overlaid on standard
single-field slow-roll dynamics, our simple setup can reconcile well-known inflationary models with current experimental constraints. If overlaid on standard
single-field slow-roll dynamics, our simple setup can reconcile well-known inflationary models with current experimental constraints. Moreover, it predicts order-one
local non-Gaussianity, well within reach of missions currently collecting data, such as SPHEREx [46]. Plot obtained using PyTransport [36]. generation of observable f
loc
NL in the encore scenario do
not rely on an additional conversion mechanism, such as
a decay channel or a modulated reheating surface. generation of observable f
loc
NL in the encore scenario do
not rely on an additional conversion mechanism, such as
a decay channel or a modulated reheating surface. Both
the curvature enhancement and its nonlinear signature
stem from the same post-inflationary axion dynamics. 995π (E ∼ O(100))
Numerical
Figure 3. ns − r plane (for f = 1/π) with the numerical (lines) and analytical (markers) predictions for chaotic
(left) and Starobinsky (right) inflation with flat internal geometry, together with the 2σ confidence regions of diverse
data sets. 995π (E ∼ O(100))
Numerical
Figure 3. ns − r plane (for f = 1/π) with the numerical (lines) and analytical (markers) predictions for chaotic
(left) and Starobinsky (right) inflation with flat internal geometry, together with the 2σ confidence regions of diverse
data sets. 995π (E ∼ O(100))
Numerical
Figure 3. ns − r plane (for f = 1/π) with the numerical (lines) and analytical (markers) predictions for chaotic
(left) and Starobinsky (right) inflation with flat internal geometry, together with the 2σ confidence regions of diverse
data sets. 995π (E ∼ O(100))
Numerical
Figure 3. ns − r plane (for f = 1/π) with the numerical (lines) and analytical (markers) predictions for chaotic
(left) and Starobinsky (right) inflation with flat internal geometry, together with the 2σ confidence regions of diverse
data sets. 995π (E ∼ O(100))
Numerical
Figure 3. ns − r plane (for f = 1/π) with the numerical (lines) and analytical (markers) predictions for chaotic
(left) and Starobinsky (right) inflation with flat internal geometry, together with the 2σ confidence regions of diverse
data sets. 995π (E ∼ O(100))
Numerical
Figure 3. ns − r plane (for f = 1/π) with the numerical (lines) and analytical (markers) predictions for chaotic
(left) and Starobinsky (right) inflation with flat internal geometry, together with the 2σ confidence regions of diverse
data sets. 995π (E ∼ O(100))
Numerical
Figure 3. ns − r plane (for f = 1/π) with the numerical (lines) and analytical (markers) predictions for chaotic
(left) and Starobinsky (right) inflation with flat internal geometry, together with the 2σ confidence regions of diverse
data sets. 995π (E ∼ O(100))
Numerical
Figure 3. ns − r plane (for f = 1/π) with the numerical (lines) and analytical (markers) predictions for chaotic
(left) and Starobinsky (right) inflation with flat internal geometry, together with the 2σ confidence regions of diverse
data sets. 995π (E ∼ O(100))
Numerical
Figure 3. ns − r plane (for f = 1/π) with the numerical (lines) and analytical (markers) predictions for chaotic
(left) and Starobinsky (right) inflation with flat internal geometry, together with the 2σ confidence regions of diverse
data sets. 995π (E ∼ O(100))
Numerical
Figure 3. ns − r plane (for f = 1/π) with the numerical (lines) and analytical (markers) predictions for chaotic
(left) and Starobinsky (right) inflation with flat internal geometry, together with the 2σ confidence regions of diverse
data sets. 995π (E ∼ O(100))
Numerical
Figure 3. ns − r plane (for f = 1/π) with the numerical (lines) and analytical (markers) predictions for chaotic
(left) and Starobinsky (right) inflation with flat internal geometry, together with the 2σ confidence regions of diverse
data sets. 995π (E ∼ O(100))
Numerical
Figure 3. ns − r plane (for f = 1/π) with the numerical (lines) and analytical (markers) predictions for chaotic
(left) and Starobinsky (right) inflation with flat internal geometry, together with the 2σ confidence regions of diverse
data sets. 995π (E ∼ O(100))
Numerical
Figure 3. ns − r plane (for f = 1/π) with the numerical (lines) and analytical (markers) predictions for chaotic
(left) and Starobinsky (right) inflation with flat internal geometry, together with the 2σ confidence regions of diverse
data sets. 995π (E ∼ O(100))
Numerical
Figure 3. ns − r plane (for f = 1/π) with the numerical (lines) and analytical (markers) predictions for chaotic
(left) and Starobinsky (right) inflation with flat internal geometry, together with the 2σ confidence regions of diverse
data sets. 995π (E ∼ O(100))
Numerical
Figure 3. ns − r plane (for f = 1/π) with the numerical (lines) and analytical (markers) predictions for chaotic
(left) and Starobinsky (right) inflation with flat internal geometry, together with the 2σ confidence regions of diverse
data sets. 995π (E ∼ O(100))
Numerical
Figure 3. ns − r plane (for f = 1/π) with the numerical (lines) and analytical (markers) predictions for chaotic
(left) and Starobinsky (right) inflation with flat internal geometry, together with the 2σ confidence regions of diverse
data sets. 995π (E ∼ O(100))
Numerical
Figure 3. ns − r plane (for f = 1/π) with the numerical (lines) and analytical (markers) predictions for chaotic
(left) and Starobinsky (right) inflation with flat internal geometry, together with the 2σ confidence regions of diverse
data sets. 995π (E ∼ O(100))
Numerical
Figure 3. ns − r plane (for f = 1/π) with the numerical (lines) and analytical (markers) predictions for chaotic
(left) and Starobinsky (right) inflation with flat internal geometry, together with the 2σ confidence regions of diverse
data sets. 995π (E ∼ O(100))
Numerical
Figure 3. ns − r plane (for f = 1/π) with the numerical (lines) and analytical (markers) predictions for chaotic
(left) and Starobinsky (right) inflation with flat internal geometry, together with the 2σ confidence regions of diverse
data sets. 995π (E ∼ O(100))
Numerical
Figure 3. ns − r plane (for f = 1/π) with the numerical (lines) and analytical (markers) predictions for chaotic
(left) and Starobinsky (right) inflation with flat internal geometry, together with the 2σ confidence regions of diverse
data sets. 995π (E ∼ O(100))
Numerical
Figure 3. ns − r plane (for f = 1/π) with the numerical (lines) and analytical (markers) predictions for chaotic
(left) and Starobinsky (right) inflation with flat internal geometry, together with the 2σ confidence regions of diverse
data sets. 995π (E ∼ O(100))
Numerical
Figure 3. ns − r plane (for f = 1/π) with the numerical (lines) and analytical (markers) predictions for chaotic
(left) and Starobinsky (right) inflation with flat internal geometry, together with the 2σ confidence regions of diverse
data sets. Together
with Eq. (24), to fit recent CMB data, we find
Flat chaotic: (37)
∆θ ≲ 10−3
, 0. Together
with Eq. (24), to fit recent CMB data, we find
Flat chaotic: (37)
∆θ ≲ 10−3
, 0. Together
with Eq. (24), to fit recent CMB data, we find
Flat chaotic: (37)
∆θ ≲ 10−3
, 0. Together
with Eq. (24), to fit recent CMB data, we find
Flat chaotic: (37)
∆θ ≲ 10−3
, 0. Together
with Eq. (24), to fit recent CMB data, we find
Flat chaotic: (37)
∆θ ≲ 10−3
, 0. Together
with Eq. (24), to fit recent CMB data, we find
Flat chaotic: (37)
∆θ ≲ 10−3
, 0. Together
with Eq. (24), to fit recent CMB data, we find
Flat chaotic: (37)
∆θ ≲ 10−3
, 0. Together
with Eq. (24), to fit recent CMB data, we find
Flat chaotic: (37)
∆θ ≲ 10−3
, 0. Together
with Eq. (24), to fit recent CMB data, we find
Flat chaotic: (37)
∆θ ≲ 10−3
, 0. Together
with Eq. (24), to fit recent CMB data, we find
Flat chaotic: (37)
∆θ ≲ 10−3
, 0. Together
with Eq. (24), to fit recent CMB data, we find
Flat chaotic: (37)
∆θ ≲ 10−3
, 0. Together
with Eq. (24), to fit recent CMB data, we find
Flat chaotic: (37)
∆θ ≲ 10−3
, 0. Together
with Eq. (24), to fit recent CMB data, we find
Flat chaotic: (37)
∆θ ≲ 10−3
, 0. Together
with Eq. (24), to fit recent CMB data, we find
Flat chaotic: (37)
∆θ ≲ 10−3
, 0. Together
with Eq. (24), to fit recent CMB data, we find
Flat chaotic: (37)
∆θ ≲ 10−3
, 0. Together
with Eq. (24), to fit recent CMB data, we find
Flat chaotic: (37)
∆θ ≲ 10−3
, 0. Together
with Eq. (24), to fit recent CMB data, we find
Flat chaotic: (37)
∆θ ≲ 10−3
, 0. Together
with Eq. (24), to fit recent CMB data, we find
Flat chaotic: (37)
∆θ ≲ 10−3
, 0. Together
with Eq. (24), to fit recent CMB data, we find
Flat chaotic: (37)
∆θ ≲ 10−3
, 0. Together
with Eq. (24), to fit recent CMB data, we find
Flat chaotic: (37)
∆θ ≲ 10−3
, 0. Together
with Eq. (24), to fit recent CMB data, we find
Flat chaotic: (37)
∆θ ≲ 10−3
, 0. Together
with Eq. (24), to fit recent CMB data, we find
Flat chaotic: (37)
∆θ ≲ 10−3
, 0. These bounds
can be translated onto more fundamental parameters via
mθ ∼ Λ
2/f. While our mechanism is general, throughout this letter we will illustrate it with the example of chaotic inflation [35] (and flat field space geometry) and, at the end,
also show the observational consequences in the Starobinsky case. These bounds
can be translated onto more fundamental parameters via
mθ ∼ Λ
2/f. While our mechanism is general, throughout this letter we will illustrate it with the example of chaotic inflation [35] (and flat field space geometry) and, at the end,
also show the observational consequences in the Starobinsky case. These bounds
can be translated onto more fundamental parameters via
mθ ∼ Λ
2/f. While our mechanism is general, throughout this letter we will illustrate it with the example of chaotic inflation [35] (and flat field space geometry) and, at the end,
also show the observational consequences in the Starobinsky case. These bounds
can be translated onto more fundamental parameters via
mθ ∼ Λ
2/f. While our mechanism is general, throughout this letter we will illustrate it with the example of chaotic inflation [35] (and flat field space geometry) and, at the end,
also show the observational consequences in the Starobinsky case. These bounds
can be translated onto more fundamental parameters via
mθ ∼ Λ
2/f. While our mechanism is general, throughout this letter we will illustrate it with the example of chaotic inflation [35] (and flat field space geometry) and, at the end,
also show the observational consequences in the Starobinsky case. These bounds
can be translated onto more fundamental parameters via
mθ ∼ Λ
2/f. While our mechanism is general, throughout this letter we will illustrate it with the example of chaotic inflation [35] (and flat field space geometry) and, at the end,
also show the observational consequences in the Starobinsky case. These bounds
can be translated onto more fundamental parameters via
mθ ∼ Λ
2/f. While our mechanism is general, throughout this letter we will illustrate it with the example of chaotic inflation [35] (and flat field space geometry) and, at the end,
also show the observational consequences in the Starobinsky case. These bounds
can be translated onto more fundamental parameters via
mθ ∼ Λ
2/f. While our mechanism is general, throughout this letter we will illustrate it with the example of chaotic inflation [35] (and flat field space geometry) and, at the end,
also show the observational consequences in the Starobinsky case. These bounds
can be translated onto more fundamental parameters via
mθ ∼ Λ
2/f. While our mechanism is general, throughout this letter we will illustrate it with the example of chaotic inflation [35] (and flat field space geometry) and, at the end,
also show the observational consequences in the Starobinsky case. These bounds
can be translated onto more fundamental parameters via
mθ ∼ Λ
2/f. While our mechanism is general, throughout this letter we will illustrate it with the example of chaotic inflation [35] (and flat field space geometry) and, at the end,
also show the observational consequences in the Starobinsky case. These bounds
can be translated onto more fundamental parameters via
mθ ∼ Λ
2/f. While our mechanism is general, throughout this letter we will illustrate it with the example of chaotic inflation [35] (and flat field space geometry) and, at the end,
also show the observational consequences in the Starobinsky case. These bounds
can be translated onto more fundamental parameters via
mθ ∼ Λ
2/f. While our mechanism is general, throughout this letter we will illustrate it with the example of chaotic inflation [35] (and flat field space geometry) and, at the end,
also show the observational consequences in the Starobinsky case. These bounds
can be translated onto more fundamental parameters via
mθ ∼ Λ
2/f. While our mechanism is general, throughout this letter we will illustrate it with the example of chaotic inflation [35] (and flat field space geometry) and, at the end,
also show the observational consequences in the Starobinsky case. These bounds
can be translated onto more fundamental parameters via
mθ ∼ Λ
2/f. While our mechanism is general, throughout this letter we will illustrate it with the example of chaotic inflation [35] (and flat field space geometry) and, at the end,
also show the observational consequences in the Starobinsky case. These bounds
can be translated onto more fundamental parameters via
mθ ∼ Λ
2/f. While our mechanism is general, throughout this letter we will illustrate it with the example of chaotic inflation [35] (and flat field space geometry) and, at the end,
also show the observational consequences in the Starobinsky case. These bounds
can be translated onto more fundamental parameters via
mθ ∼ Λ
2/f. While our mechanism is general, throughout this letter we will illustrate it with the example of chaotic inflation [35] (and flat field space geometry) and, at the end,
also show the observational consequences in the Starobinsky case. These bounds
can be translated onto more fundamental parameters via
mθ ∼ Λ
2/f. While our mechanism is general, throughout this letter we will illustrate it with the example of chaotic inflation [35] (and flat field space geometry) and, at the end,
also show the observational consequences in the Starobinsky case. These bounds
can be translated onto more fundamental parameters via
mθ ∼ Λ
2/f. While our mechanism is general, throughout this letter we will illustrate it with the example of chaotic inflation [35] (and flat field space geometry) and, at the end,
also show the observational consequences in the Starobinsky case. These bounds
can be translated onto more fundamental parameters via
mθ ∼ Λ
2/f. While our mechanism is general, throughout this letter we will illustrate it with the example of chaotic inflation [35] (and flat field space geometry) and, at the end,
also show the observational consequences in the Starobinsky case. These bounds
can be translated onto more fundamental parameters via
mθ ∼ Λ
2/f. While our mechanism is general, throughout this letter we will illustrate it with the example of chaotic inflation [35] (and flat field space geometry) and, at the end,
also show the observational consequences in the Starobinsky case. These bounds
can be translated onto more fundamental parameters via
mθ ∼ Λ
2/f. While our mechanism is general, throughout this letter we will illustrate it with the example of chaotic inflation [35] (and flat field space geometry) and, at the end,
also show the observational consequences in the Starobinsky case. These bounds
can be translated onto more fundamental parameters via
mθ ∼ Λ
2/f. While our mechanism is general, throughout this letter we will illustrate it with the example of chaotic inflation [35] (and flat field space geometry) and, at the end,
also show the observational consequences in the Starobinsky case.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Enhanced Cosmological Model Reconciliation: The improved AI system can reconcile otherwise disfavored single-field inflationary models with current data by incorporating a light axion spectator field that reshapes inflationary observables through purely gravitational multi-field dynamics.
-
Prediction of Local Non-Gaussianity: The system can predict local non-Gaussianity, f loc, on the order of O(1), which is within reach of upcoming surveys like SPHEREx.
-
Parameter Space Mapping for Inflationary Models: The AI can map the relationship between model parameters such as m2 theta/m2 phi and Lambda4/V0 to determine the required range for axion decay constant f, enabling it to constrain viable inflationary scenarios based on CMB data.
-
Spectral Tilt Determination: The system can calculate the post-inflationary spectral tilt (n post s) using equations like Eq. (19) and Eq. (22), allowing it to differentiate between standard single-field results and those modified by the encore mechanism, especially in the E >> 1 regime where n post s is dictated mostly by the entropic mode.
-
Tensor-to-Scalar Ratio Suppression: The system can calculate a suppressed tensor-to-scalar ratio r post using Eq. (18), providing a quantitative measure of how the spectator axion mechanism reduces r compared to the single-field case, which is crucial for testing gravitational wave constraints.
Abstract
We propose a generic mechanism applicable to single-field inflaton models in which a light axion spectator, initialized near the hilltop of its potential, reshapes inflationary observables through purely gravitational multi-field dynamics. During inflation the axion is frozen and the background follows an effectively single-field trajectory. After inflation ends, the axion rolls, inducing a turn in field space and transient tachyonic phases of the isocurvature mode. The resulting ``tachyonic encore'' occurs entirely on super-horizon scales. These phases generate a nearly scale-invariant enhancement of the curvature power spectrum, suppressing the tensor-to-scalar ratio and shifting the scalar tilt to a weighted combination of adiabatic and entropic tilts at horizon crossing. We show that these effects can reconcile otherwise disfavored inflaton potentials with current CMB constraints. The same dynamics predict local non-Gaussianity, f NL loc. about O(1), within reach of upcoming surveys.
Sources
- Particle Physics and Inflationary Cosmology
- Inflationary Cosmology
- Inflation: Theory and Observations
- Inflation (2025)
- On the Present Status of Inflationary Cosmology
- Planck 2018 results. X. Constraints on inflation
- BICEP / Keck XIII: Improved Constraints on Primordial Gravitational Waves using Planck, WMAP, and BICEP/Keck Observations through the 2018 Observing Season
- The Atacama Cosmology Telescope: DR6 Power Spectra, Likelihoods and $\Lambda$CDM Parameters
- String Axiverse
- The type IIB string axiverse and its low-energy phenomenology
- The Kreuzer-Skarke Axiverse
- Glimmers from the Axiverse
- The COSMIC WISPers White Paper: The physics case for Weakly Interacting Slim Particles
- WISPedia -- the WISPs Encyclopedia
- Naturally inflating on steep potentials through electromagnetic dissipation
- Stability analysis of chromo-natural inflation and possible evasion of Lyth's bound
- Scale-dependent gravitational waves from a rolling axion
- Rolling axions during inflation: perturbativity and signatures
- Double Monodromy Inflation: A Gravity Waves Factory for CMB-S4, LiteBIRD and LISA
- Very Hairy Inflation
Related papers
- The facts on quantum self-dual Einstein gravity and self-dual Yang-Mills theory
- Quantum Langlands correspondence for Painlev'e II
- Fortuitous BMN Operators from Multiple Cores
- Building far-from-equilibrium effective field theories using shift symmetries
- Inflation, Cosmic Perturbations and Non-Gaussianities
- Wavefunctions of AdS 3 Universes and T-deformed Torus Partition Functions