{"id":"e8d9b504-5e80-415a-ad3e-56af9147dd9b","arxiv_id":"1403.5883","paper_version":2,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A two-sine-wave inflaton potential can span a broad region of (n_s, r) space, including n_s near or above unity with r around 0.1 to 0.2, so the model is not excluded by BICEP2.","lead":"Multi-natural inflation, a model where the inflaton rolls on a sum of sine waves, can produce a wide range of spectral index and tensor-to-scalar ratio values, including a spectral index near or above one. The authors argue this flexibility lets the model survive the 2014 BICEP2 measurement even if the BICEP2-Planck tension is resolved by something other than running.","discovery_kind":"unclear","skeptic_critique":{"model":"selfhosted/deepseek-v4.1-flash","headline":"Fig. 1's wide (n_s, r) range — including n_s ≥ 1 — is shown only for θ = 2π/3 and f2/f1 = 0.5; the abstract attributes this breadth to 'two sinusoidal functions' generally, but no scan over θ or f2/f1 supports that attribution.","rationale":"The reader's weakest_assumption identifies the right load-bearing concern: the Fig. 1 scan varies only f while fixing B, f2/f1, and θ at hand-picked values, so the claimed breadth of (n_s, r) is not established as a property of the model's parameter space. I agree, and I sharpen it: the parameter that most directly controls whether n_s ≥ 1 is reached is θ, because it sets the sign and location of the V'' > 0 region relative to where N = 60 lands. The abstract's phrasing ('multi-natural inflation can accommodate a wide range') attributes the breadth to the two-sine structure generically, which the fixed-θ demonstration does not support. That is the single most load-bearing soft spot. Two of the reader's other points are weaker than the reader suggests. The running spectral index is explicitly scoped out in Section III ('we focus on the case without running to simplify our analysis') and the abstract attributes it to prior work ('it is known'), so it is an acknowledged scope limitation rather than a hidden flaw; per the citation rule it should not be treated as load-bearing. The string construction in Section II D is explicitly labeled 'toward a UV completion' and is not load-bearing for the central claim. The supergravity derivation in Sections II A–B (Kähler potential, saxion stabilization via ∂_Φ K = 0, axion potential Eq. 16, decoupling under condition (7)) is standard and I found no error there. The numerical consistency of the quoted numbers also checks out: with r = 0.2, ε ≈ 0.0125 and −6ε ≈ −0.075, so n_s ≈ 1 requires η ≈ 0.04, which is achievable for f1 ≈ 4 Mpl where 1/f1² ≈ 0.06. So the physics hangs together; the open question is representativeness of the slice, not correctness of the arithmetic. The reader's CONDITIONAL verdict is appropriate and my sharpening does not move it, so UNCHANGED. The concrete test above (a θ × f2/f1 grid) would settle whether the concern lands.","tokens_in":12823,"tokens_out":8274,"duration_ms":175576,"concrete_test":"Recompute Fig. 1 on a grid of θ ∈ [0, 2π) (say 24 values) and f2/f1 ∈ {0.3, 0.4, 0.5, 0.6, 0.7}, with B ∈ {0.1, 0.2, 0.3, 0.4}, scanning f1 as in Section III and imposing N = 60. For each point record (n_s, r) and whether the trajectory completes 60 e-folds without being trapped. If the region n_s ≥ 1 with r = 0.1–0.2 appears only for θ within a narrow window (say |θ − 2π/3| < π/6) and f2/f1 = 0.5, the 'wide range' claim is a slice artifact and the abstract's attribution to 'two sinusoidal functions' is unsupported. If it persists over a substantial fraction of the grid, the concern does not land.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim (abstract:s2) is that multi-natural inflation 'can accommodate a wide range of values of (n_s, r), including the spectral index close to or even above unity.' The evidence is Fig. 1, computed in Section III with Eq. 34 fixing f2 = 0.5 f1, θ = 2π/3, and B ∈ {0.20, 0.25, 0.30}, scanning only f. The concern: n_s ≥ 1 requires η = Mpl² V''/V > 0 at horizon crossing, i.e., the field sitting on a convex (V'' > 0) stretch of the modulated potential. With f2 = 0.5 f1 the second term is the second harmonic, V(x) = C − Λ1⁴[cos x + B cos(2x + θ)], and the sign and location of the V'' > 0 regions are controlled by θ. At θ = 0 the potential is symmetric and the field can be trapped at x = 0; θ = 2π/3 supplies a specific tilt that both permits rolling and places the convex region where N = 60 lands. A different θ (or f2/f1 ≠ 0.5) shifts where horizon crossing occurs relative to the convex stretch and can move n_s below unity or change r. The paper gives no argument that θ = 2π/3 and f2/f1 = 0.5 are representative, yet the abstract's phrasing attributes the breadth to the two-sine structure itself. If n_s ≥ 1 survives only in a narrow window around θ = 2π/3, the 'wide range' is a property of a fine-tuned slice, not of the model. This is the reader's concern, sharpened to the specific parameter (θ) that controls the sign of η.","agreement_with_reader":"agree"},"referee_report":{"model":"selfhosted/deepseek-v4.1-flash","summary":"The manuscript revisits multi-natural inflation — a single-field model in which the inflaton potential is a sum of two sinusoidal functions with different heights and periodicities — and its realization in supergravity, in light of the BICEP2 detection of primordial B-modes. After reviewing the saxion stabilization and the axion potential in Section II, the authors present a string-inspired embedding with several Kähler moduli and non-perturbative terms (Section II.D), and then, in Section III, numerically scan the decay constant f for three fixed values of the amplitude ratio B = Λ2⁴/Λ1⁴, with f2 = 0.5 f1 and θ = 2π/3 held fixed (Eq. 34). The resulting Fig. 1 is used to argue that the model can accommodate a wide range of (n_s, r), including n_s close to or above unity together with r ≈ 0.1–0.2, which the authors note would be favored if the BICEP2/Planck tension is resolved by dark radiation, hot dark matter, or non-zero neutrino mass rather than by a running spectral index.","tokens_in":13258,"tokens_out":5465,"duration_ms":99950,"significance":"If the existence claim holds, the paper is a useful addition to the large-field inflation model space: it offers an explicit supergravity construction with a controlled saxion, a plausible string-theoretic embedding, and a falsifiable prediction in the (n_s, r) plane that future CMB and 21 cm observations can test. The supergravity derivation in Section II is explicit and checkable, and the string embedding in Section II.D is a genuine attempt at UV completion rather than a hand-waving gesture. The paper is a follow-up to the authors' earlier work (Refs. [21], [22], [35]); the new element is the (n_s, r) scan of Section III, and the manuscript is appropriately explicit that the running spectral index case is treated elsewhere.","major_comments":[{"comment":"The numerical scan fixes θ = 2π/3, f2 = 0.5 f1, and B ∈ {0.20, 0.25, 0.30}, varying only f. The abstract attributes the breadth of the (n_s, r) region to the two-sinusoidal-function structure in general, but θ controls the sign and location of the V'' > 0 stretches of the modulated potential and therefore whether n_s ≥ 1 is reachable at N = 50–60. Please add a scan over θ (and, ideally, over f2/f1) showing the range of (n_s, r) obtained, or state explicitly in Section III and in the abstract that Fig. 1 is a representative slice at θ = 2π/3 and f2 = 0.5 f1. This is the load-bearing point for the headline claim.","section":"Section III, Eq. (34) and Fig. 1","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The claim that n_s 'close to or even above unity' is realized is not quantitatively checkable from the figure as printed: the curves are not labeled with the corresponding values of f, and the n_s axis extends only to 1.02. Please tabulate (n_s, r) at the endpoints of each curve for B = 0.20, 0.25, 0.30, and state the value of f at which n_s = 1 is crossed, if it is crossed at all. Without this, the reader cannot verify whether the n_s ≥ 1 region is reached or merely approached.","section":"Section III, Fig. 1","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The string embedding is schematic in places. The estimate f ∼ 1/((t1+t2)^{1/4} V^{1/2}) ∼ 0.1 and the claim that f1 and f2 can be super-Planckian for n1 ≳ n2 = O(10) with Δ1 = O(0.1) and Δ2 = O(1) should be supported by a short derivation or an explicit reference, and the condition under which the cross terms between shift-symmetry-breaking exponentials can be neglected in Eq. (22) should be stated. As written, the reader must reconstruct these steps.","section":"Section II.D, Eqs. (26)–(32)","authors_acknowledge":null,"authors_acknowledge_p":null}],"minor_comments":[{"comment":"The phrase 'This will be be favored' contains a duplicated word and should read 'This will be favored'.","section":"Abstract","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"'Substituing' should be 'Substituting'.","section":"Section II.B, below Eq. (16)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The coefficient (2/(f1 f2) − 3) of the cross term and the condition under which the third term is negligible follow from Eq. (7), but a sentence spelling out the parametric suppression (AB/(AW0) ∼ B/W0 ≪ 1) would help the reader see why the two-sine potential of Ref. [21] is recovered.","section":"Section II.B, Eq. (16)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The exponent in Ae^{−2π/(8n1)(8T2+9T4)} is ambiguous as typeset; please write it with explicit parentheses so that the reader can see whether the factor 8n1 multiplies the whole combination 8T2 + 9T4.","section":"Section II.D, Eq. (32)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The saxion mass mσ ≃ √2|W0| should be stated explicitly as being in reduced Planck units, consistent with the convention declared below Eq. (7).","section":"Section II.A, Eq. (14)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The labels 3.5Mpl, 4.0Mpl, etc. on the brown dashed curve should be identified in the caption as values of f, and the brown dashed curve itself should be identified as the natural-inflation prediction, not left to the reader to infer from the legend.","section":"Fig. 1 caption","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"Since the abstract emphasizes the running spectral index, a sentence connecting Fig. 1 (computed without running) to the running case of Ref. [35] would help the reader understand which part of the abstract's motivation is being tested here and which is taken from earlier work.","section":"Section III, opening paragraph","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The reference entry 'arXiv:1310.1774 [hep-ph].' is missing a closing bracket and should read 'arXiv:1310.1774 [hep-ph]]'.","section":"References, [30]","authors_acknowledge":null,"authors_acknowledge_p":null}],"recommendation":"minor_revision","confidential_remarks":"The manuscript is a follow-up to the authors' Refs. [21], [22], and [35], and the supergravity setup of Section II overlaps substantially with [22]. The novel content is the (n_s, r) scan of Section III. The editor may wish to confirm that the novelty relative to [21] and [22] is stated clearly enough in the introduction for readers who have not followed the series."},"author_rebuttal":null,"desk_editor":{"model":"selfhosted/deepseek-v4.1-flash","letter":"This is the third paper in the authors' multi-natural inflation sequence, and it does what a follow-up should: takes the model from Refs. [21] and [22] and asks what BICEP2 does to it. The genuinely new pieces are the numerical scan in Fig. 1 showing that the two-sine potential V = C − Λ₁⁴cos(φ/f₁) − Λ₂⁴cos(φ/f₂ + θ) can land at n_s close to or even above unity with r ≈ 0.1–0.2, and the five-modulus string construction in Sec. II D. The supergravity derivation in Sec. II is standard and clean; the saxion/axion hierarchy argument (H_inf ∼ √(AW₀) ≪ W₀, so the saxion decouples) is a nice, correct point and worth having in the literature. The string section is explicitly labelled 'toward a UV completion' and is not checked for consistency — that is honest, and I would not hold it against them, but it also should not be read as a construction that works. The soft spot is the one your stress-test note points at, and it lands. Eq. 34 fixes f₂ = 0.5 f₁, θ = 2π/3, and B ∈ {0.20, 0.25, 0.30}; only f is scanned. Since θ controls where the convex (V'' > 0) stretch sits relative to horizon crossing, n_s ≥ 1 plausibly survives only in a window around θ = 2π/3. The abstract's 'wide range' then describes breadth in f, not breadth in the model's parameter space. That said, the paper's claim is 'can accommodate', an existence statement, and one working slice is enough for that. The problem is phrasing, not arithmetic. A scan over θ at fixed f, or even a sentence saying why 2π/3 is generic, would settle it. Second, smaller point: the running spectral index that leads the abstract is imported from Ref. [35], a companion preprint by overlapping authors, and is not computed here. That is fine as a citation but it should not carry the abstract's weight. Also worth saying plainly: the BICEP2 r = 0.2 that motivates the whole exercise has since gone away, so the phenomenological hook is dated even though the model survives. Who should read it: anyone working on axion inflation, aligned/hybrid axion potentials, or moduli stabilization with multiple non-perturbative terms. It is a useful reference for the (n_s, r) reach of two-sine potentials. I would cite it for that. It deserves a referee, but a light one — the physics is sound and the fixes are editorial.","headline":"A modest, honest follow-up: the two-sine axion potential can reach n_s ≳ 1 with r ≈ 0.2, but the 'wide range' rests on one hand-picked slice of parameter space.","tokens_in":13899,"tokens_out":1338,"would_cite":true,"duration_ms":28245,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"pith_extraction":{"msc":[],"pacs":[],"model":"selfhosted/deepseek-v4.1-flash","headline":"A two-sinusoid inflaton potential can push the scalar spectral index to one, or above, while keeping a large tensor-to-scalar ratio.","keywords":["multi-natural inflation","axion inflation","supergravity","spectral index","tensor-to-scalar ratio","BICEP2","moduli stabilization","string compactification"],"falsifier":"A joint CMB constraint on $n_s$, $r$, and $dn_s/d$ln$k$ that lands at $n_s$ clearly below unity with $r$ near $0.2$, no running, and no extra radiation or hot dark matter: that would remove the motivation for the high-$n_s$ branch and leave the paper's distinctive region unoccupied. Conversely, a measurement of $n_s$ above unity at several sigma together with $r$ around $0.15$ would single out the multi-sine interference region the paper plots.","tokens_in":14187,"feed_emoji":"🌌","tokens_out":3375,"duration_ms":68212,"temperature":0.7,"pith_summary":"The BICEP2 detection of primordial B-modes, read at face value, points to large-field inflation with a tensor-to-scalar ratio around $r=0.2$, but it sits in tension with Planck's preference for a slightly red spectral index. This paper asks whether a single-field model whose potential is a sum of two sine functions of the inflaton — multi-natural inflation — can cover enough of the $(n_s, r)$ plane to absorb that tension, either through a running spectral index or through a spectral index near or above unity once the tension is blamed on something else, such as extra radiation, hot dark matter, or neutrino mass. The authors also show how the same potential arises in supergravity from the axion of a modulus field, with the saxion stabilized heavy at the origin, and sketch a string-theoretic embedding in which non-perturbative effects on Kähler moduli supply the two sine terms. The payoff, if the construction holds, is that a model built from a symmetry rather than from a hand-shaped potential can sit anywhere along a broad stretch of the observationally allowed band.","feed_headline":"Two sine waves let inflation's spectral index reach one","feed_subtitle":"Varying one decay constant sweeps $r$ from 0.1 to 0.2 and $n_s$ from 0.94 to above unity in a single-field model.","key_machinery":"The load-bearing object is the axion of a modulus chiral superfield, protected by a shift symmetry $φ → φ + C$ that is broken explicitly by two exponentials in the superpotential, $W = W_0 + A e^{-aΦ} + B e^{-bΦ}$, with $K = (f^2/2)(Φ + Φ^†)^2$. The Kähler potential's shift symmetry keeps the axion light while the saxion is stabilized at the origin by $∂_Φ K = 0$ and given a mass of order $|W_0|$, heavier than the Hubble scale during inflation $H_{inf} ∼ √(A W_0)$; that hierarchy is what reduces a two-field system to an effectively single-field inflation model. The two exponentials generate two sine terms with decay constants $f_1 = √2 f/a$ and $f_2 = √2 f/b$ and different heights, and it is the interference between them that bends the $(n_s, r)$ trajectory away from ordinary natural inflation. For super-Planckian decay constants without tuning, the authors use a second construction with two axions, where a heavy combination is fixed at the origin by a steeper exponential and the light combination inherits an effectively enlarged decay constant $f_1 = 2aΔ_1 f$ through accidental cancellation between terms — the alignment mechanism. A string-theoretic version replaces the exponentials with gaugino-condensation terms on a Calabi–Yau space with five Kähler moduli, arranged so that three combinations are fixed supersymmetrically and two light axion combinations drive inflation.","core_discovery":"The paper's central claim is that multi-natural inflation — a single-field model with $V(φ) = C - Λ_1^4 cos(φ/f_1) - Λ_2^4 cos(φ/f_2 + θ)$ — can produce a spectral index close to or even above unity together with $r = 0.1$–$0.2$, matching the BICEP2 range. The authors numerically solve the inflaton dynamics for $f_2 = 0.5 f_1$, $θ = 2π/3$, and three amplitude ratios $B = Λ_2^4/Λ_1^4 = 0.20, 0.25, 0.30$, varying the decay constant $f$ across the range where the field excursion reaches roughly $3.5$ to $5$ Planck masses. The resulting trajectories in the $(n_s, r)$ plane sweep from about $n_s = 0.94$ up past $n_s = 1.00$ at $r$ between roughly $0.1$ and $0.2$, so the model spans both the Planck-favoured region and the region that would be preferred if the BICEP2–Planck tension were resolved by dark radiation, hot dark matter, or non-zero neutrino mass. This is presented as a distinct virtue from the model's previously noted ability to generate a large running spectral index from small modulations of the potential; here the authors deliberately switch the running off to isolate what the multiple sine terms do to $(n_s, r)$ on their own.","pith_inferences":["The predicted $(n_s, r)$ relation is a one-parameter curve for each choice of amplitude ratio and phase, so a joint measurement of $n_s$ and $r$ with modest error bars would over-determine the model and could confirm or exclude the specific slice $f_2 = 0.5 f_1$, $θ = 2π/3$ that the paper plots.","A detection of $n_s$ measurably above unity would be a sharper discriminator than $r$ alone: it is hard to reach in single-sine natural inflation, and it would point specifically at the multi-sine interference term rather than at the overall scale of the potential.","The running and the large-$n_s$ branches are in principle separable by future 21cm or CMB data, since they predict different signs and magnitudes for $dn_s/d\\ln k$ at the same $r$; that separation is a cleaner test than fitting either alone.","The alignment mechanism used here to get super-Planckian decay constants is generic, so the same two-axion arrangement could be borrowed by other large-field models that need a super-Planckian scale without trans-Planckian input parameters."],"forward_implications":["If the BICEP2–Planck tension is resolved by extra radiation, hot dark matter, or neutrino mass rather than by running, the data should settle near $n_s$ close to or above unity with $r \\approx 0.1$–$0.2$, a corner that multi-natural inflation reaches and ordinary natural inflation does not.","The model keeps its earlier mechanism for a sizable running spectral index, since the mild hierarchy $f_2 \\sim 10^{-2} f_1$ needed for running is a built-in feature of the same two-sine potential rather than an added ingredient.","Because the saxion mass exceeds the Hubble scale during inflation, the moduli destabilization problem that afflicts many large-field supergravity models is avoided in this construction, so the model can be embedded in a moduli-stabilized compactification.","The tensor amplitude fixes the scale of the shift-symmetry-breaking terms, $A W_0 = O(10^{-9})$ for $f_1 = O(10)$, tying a CMB observable directly to superpotential parameters.","The two-axion alignment construction allows super-Planckian effective decay constants with $f \\sim 0.1$ and order-one exponents, so the model does not require trans-Planckian input parameters."],"supporting_citations":[{"why":"Supplies the BICEP2 B-mode detection, the $r$ range, and the stated tension with Planck that the whole paper responds to.","marker":"[1]"},{"why":"Proposes multi-natural inflation, the two-sinusoid single-field potential that this paper extends and re-scans.","marker":"[21]"},{"why":"Gives the supergravity realization with an axion modulus, saxion stabilization, and the $A W_0$ scale that this paper builds on.","marker":"[22]"},{"why":"Establishes that multi-natural inflation generates a sizable running spectral index from small modulations, the alternative route to the same tension.","marker":"[35]"},{"why":"Shows that small modulations of an inflaton potential produce large running without spoiling the background dynamics, the mechanism behind the running branch.","marker":"[36]"},{"why":"Supplies the two-axion alignment idea by which an effectively super-Planckian decay constant arises from cancellation between potentials.","marker":"[47]"},{"why":"Provides the analysis showing that dark radiation, hot dark matter, or neutrino mass can relieve the BICEP2–Planck tension, the scenario in which $n_s$ near unity is favoured.","marker":"[24]"},{"why":"Supplies the moduli-stabilization framework used in the string-theoretic embedding of the inflationary sector.","marker":"[40]"},{"why":"Gives the field-excursion bound linking a large $r$ to super-Planckian inflaton displacement, the reason a shift symmetry is needed at all.","marker":"[32]"}],"fun_headline_variants":["Two sine terms let inflation's spectral index reach unity","Multi-sine inflaton spans BICEP2 r and n_s above one","Single-field sine potential reaches n_s close to one","Sine-wave inflation covers BICEP2 band, n_s near one","Multi-natural inflation can push n_s to one and higher"],"cache_read_input_tokens":0,"weakest_assumption_plain":"The claim that a wide range of $(n_s, r)$ is reachable rests on a scan that varies only the overall decay constant while holding the amplitude ratio at three hand-picked values, the decay-constant ratio at $0.5$, and the phase at $2π/3$, with no argument that this slice represents what the model can do elsewhere in its parameter space.","fun_headline_variants_meta":{"raw":{"variants":["Two sine terms let inflation's spectral index reach unity","Multi-sine inflaton spans BICEP2 r and n_s above one","Single-field sine potential reaches n_s close to one","Sine-wave inflation covers BICEP2 band, n_s near one","Multi-natural inflation can push n_s to one and higher"]},"model":"selfhosted/deepseek-v4.1-flash","effort":"low","attempts":1,"cost_usd":0.0,"raw_usage":{"total_tokens":1565,"prompt_tokens":1001,"completion_tokens":564,"prompt_tokens_details":null,"completion_tokens_details":{"reasoning_tokens":489}},"tokens_in":1001,"tokens_out":564,"duration_ms":10294,"image_pages":[],"temperature":1.0,"reasoning_budget":50,"reasoning_tokens":489,"transport_provider":"selfhosted","cache_read_input_tokens":0,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-11-10T08:09:04.926244+00:00","model_set":{"reader":"selfhosted/deepseek-v4.1-flash"},"falsifier":"A joint CMB constraint on $n_s$, $r$, and $dn_s/d$ln$k$ that lands at $n_s$ clearly below unity with $r$ near $0.2$, no running, and no extra radiation or hot dark matter: that would remove the motivation for the high-$n_s$ branch and leave the paper's distinctive region unoccupied. Conversely, a measurement of $n_s$ above unity at several sigma together with $r$ around $0.15$ would single out the multi-sine interference region the paper plots.","supporting_citations":[{"cited_title":"BICEP2 I: Detection Of B-mode Polarization at Degree Angular Scales","cited_arxiv_id":"1403.3985","evidence_quote":"Supplies the BICEP2 B-mode detection, the $r$ range, and the stated tension with Planck that the whole paper responds to."},{"cited_title":"Multi-Natural Inflation","cited_arxiv_id":"1401.5212","evidence_quote":"Proposes multi-natural inflation, the two-sinusoid single-field potential that this paper extends and re-scans."},{"cited_title":"Multi-Natural Inflation in Supergravity","cited_arxiv_id":"1403.0410","evidence_quote":"Gives the supergravity realization with an axion modulus, saxion stabilization, and the $A W_0$ scale that this paper builds on."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes that multi-natural inflation generates a sizable running spectral index from small modulations, the alternative route to the same tension."},{"cited_title":"Running Spectral Index from Inflation with Modulations","cited_arxiv_id":"1011.3988","evidence_quote":"Shows that small modulations of an inflaton potential produce large running without spoiling the background dynamics, the mechanism behind the running branch."},{"cited_title":"Completing Natural Inflation","cited_arxiv_id":"hep-ph/0409138","evidence_quote":"Supplies the two-axion alignment idea by which an effectively super-Planckian decay constant arises from cancellation between potentials."},{"cited_title":"Relic Neutrinos, thermal axions and cosmology in early 2014","cited_arxiv_id":"1403.4852","evidence_quote":"Provides the analysis showing that dark radiation, hot dark matter, or neutrino mass can relieve the BICEP2–Planck tension, the scenario in which $n_s$ near unity is favoured."},{"cited_title":"de Sitter Vacua in String Theory","cited_arxiv_id":"hep-th/0301240","evidence_quote":"Supplies the moduli-stabilization framework used in the string-theoretic embedding of the inflationary sector."},{"cited_title":"What Would We Learn by Detecting a Gravitational Wave Signal in the Cosmic Microwave Background Anisotropy?","cited_arxiv_id":"hep-ph/9606387","evidence_quote":"Gives the field-excursion bound linking a large $r$ to super-Planckian inflaton displacement, the reason a shift symmetry is needed at all."}],"review_version":1}