{"id":"e1f2a9c3-4f31-461a-985e-30cd7cbb7fa4","arxiv_id":"2411.09992","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":3.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"For f(R)=R+αR^n, the dual scalar-tensor potential is bounded below with a minimum exactly for even n and positive α, matching the f''(R)>0 condition.","lead":"This paper asks whether the stability condition of f(R) gravity, usually written as f'(R)>0 and f''(R)>0, is preserved when the theory is mapped by a conformal transformation to an equivalent scalar-tensor theory. For the polynomial models R+αR^n and R+αR^n+βR^m, the authors find that the Einstein-frame scalar potential is stable exactly when the f(R) theory is Ostrogradsky-stable.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 'stable' branch for even n>2 has a cuspy potential minimum with divergent V'', so the claimed scalar-sector stability and one-to-one correspondence are not established.","rationale":"The reader's weakest assumption identifies exactly the cuspy minimum for even n>2 and the absence of a perturbation analysis. My stress-test agrees: this is the most load-bearing weakness because the central claim is a one-to-one equivalence between scalar-sector stability and f''(R)>0. For the polynomial models the paper marks stable, the minimum at σ=0 is non-analytic with divergent second derivative, so the standard notion of perturbative stability is not satisfied. The paper's own criterion (real, bounded below, at least one minimum) is necessary but not sufficient for a stable scalar field theory. Moreover, the field redefinition between R and σ is singular at R=0 for n>2, which casts doubt on whether the Einstein-frame action is a valid dual description at the claimed vacuum. The paper also overreaches from polynomial examples to general higher-curvature theories, but that is secondary once the stability criterion itself is inadequate. The reader's conditional verdict is appropriate: the correspondence may survive if the authors add a proper perturbation analysis or explicitly adopt a non-perturbative notion of stability, but as written the central claim is not fully established. I therefore recommend no change to the verdict.","tokens_in":7804,"tokens_out":14871,"duration_ms":159318,"concrete_test":"Take f(R)=R+αR^4, α>0, in 4d. Expand the Einstein-frame potential V(σ)=3[(e^{-σ}-1)/4]^{4/3}e^{2σ} around σ=0. Show V(σ)=C|σ|^{4/3}+O(σ^{2}) with C>0, so V''(0)=∞. Then compute the quadratic part of the perturbation action δS=∫(δσ□δσ - V''(0)δσ²); it is undefined. If instead one regularizes with a small f''(0)=ε, the limit ε→0 changes the spectrum discontinuously. This single check decides whether the 'stable' n=4 branch admits a linearized scalar sector.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The classification of even n>2, α>0 as stable rests on a stability criterion that is too weak to support the claimed one-to-one correspondence. For f(R)=R+αR^n with n even and n>2, the conformal scalar is defined by f'(R)=e^{-σ}; because f''(0)=0, the map R↦σ is not a local diffeomorphism at the would-be vacuum. The resulting potential behaves as V(σ)∝|σ|^{n/(n-1)} near σ=0, e.g. |σ|^{4/3} for n=4, so V''(0) diverges and the quadratic (linearized) perturbation theory around the claimed minimum is undefined. Section 3 declares stability to be 'real, at least one minimum, bounded from below' and never checks the Hessian or the validity of the field redefinition. Therefore the paper has not shown that f(R)=R+αR^4 is perturbatively stable in the dual frame, and the asserted equivalence with f''(R)>0 (which is only nonnegative, and zero at R=0) is not established. A second limitation is that Sections 4-9 check only finite polynomial f(R) with integer powers; Section 10 nevertheless concludes a one-to-one correspondence for 'higher curvature theories' in general.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper rewrites f(R) gravity in the Einstein frame via a conformal transformation, obtaining a scalar-tensor action with potential V(σ)=(f'(A)A-f(A))/(f'(A))^{d/(d-2)}. It proposes that the f(R) theory is stable if the dual scalar potential is real, bounded below, and has at least one minimum, and that the sign of the scalar kinetic term rules out f'(R)<0. The authors then analyze f(R)=R+αR^n and f(R)=R+αR^n+βR^m for integer powers, using plots and parametric examples, and conclude that scalar-sector stability requires even powers and positive coefficients. This is summarized as a one-to-one correspondence between scalar instability and Ostrogradsky instability. The central claim is therefore that checking the Einstein-frame potential is equivalent to checking f'(R)>0 and f''(R)>0 for these polynomial models.","tokens_in":8007,"tokens_out":13510,"duration_ms":139834,"significance":"If established, the intended result would be practically useful: it would give a quick stability diagnostic for polynomial f(R) models and would explicitly connect the conformal-frame scalar potential with the standard no-ghost/no-tachyon conditions. The paper does derive the potential and the extremum conditions (Eqs. (20)-(21)) in the right direction, and the suggestion that odd powers or negative coefficients lead to instability is plausible for the families considered. However, the manuscript currently does not provide a sufficiently rigorous stability criterion: the cuspy minima for even n>2 are not analyzed perturbatively, and the claimed general equivalence goes beyond what is proved. There is also a printed sign error in Eq. (19) and an apparent sign typo in Eq. (22). These issues prevent acceptance in the present form.","major_comments":[{"comment":"Equation (19) contains a sign error. From V=(f'A-f)/(f')^2, the derivative with respect to f' is dV/df' = -A/(f')^2 + 2f/(f')^3, not A/(f')^2 + 2f/(f')^3. With the printed sign, setting dV/dσ=0 would give Af'=-2f, contradicting Eq. (20). The correct derivative restores Eq. (20) and Eq. (21), so the final conditions survive, but the displayed equation and the immediately preceding chain-rule step must be corrected.","section":"§3, Eq. (19)"},{"comment":"The stability criterion in Section 3—real, bounded below, at least one minimum—is too weak for the central claim. For f(R)=R+αR^n with even n>2 and α>0 (Section 7.1, Eq. (30) for n=4), the conformal relation f'=e^{-σ} gives R∝(e^{-σ}-1)^{1/(n-1)}, whose derivative diverges at σ=0; consequently V(σ) behaves as |σ|^{n/(n-1)} near the claimed minimum, is C^1 but not C^2, V'' diverges, and linearized scalar perturbations are ill-defined. The paper supplies no nonlinear perturbation or stability analysis, so the classification of these models as stable is not established. Additionally, at R=0 one has f''(0)=0 for these models, so the asserted equivalence with the standard condition f''>0 is not demonstrated; Section 10's statement that f'' is 'positive for any integer p or any value of R' is inaccurate (it is non-negative, vanishing at R=0 for p>1).","section":"§3 and §7"},{"comment":"The final one-to-one correspondence is stated for 'higher curvature theories' in general, but Sections 4-9 establish at most a classification for the two polynomial families R+αR^n and R+αR^n+βR^m with integer n,m. Section 9 says that higher powers and arbitrary α,β were 'checked' without showing the analysis, and no general argument is provided. Either the conclusion should be restricted to the polynomial families with a complete proof for all n,m, or the general claim should be supported by an explicit argument.","section":"§8 and §10"},{"comment":"For f'(R)<0, Eq. (14) flips the sign of both the Ricci term and the scalar kinetic term. The paper concludes instability solely from the sign of the scalar kinetic term, but the wrong-sign R term itself indicates a ghost in the gravitational sector. The necessity of f'(R)>0 should be argued more carefully, since inspecting the kinetic term in an action whose gravitational part is non-standard is not a complete stability test.","section":"§2, Eq. (14)"}],"minor_comments":[{"comment":"In Eq. (22), the factor e^{-σ d/2} appears to be a typo; from Eq. (10) and f'=e^{(2-d)σ/2} the prefactor should be e^{+σ d/2}. For d=4 this gives e^{2σ}, which is the factor used in Eqs. (23)-(31).","section":"§4, Eq. (22)"},{"comment":"The spelling 'Ostragadsky' should be 'Ostrogradsky', and headings such as 'Stablility analysis' should be corrected to 'Stability analysis'. There are also minor grammatical slips ('It there any one to one correspondence?', 'minima' used as singular).","section":"Throughout"},{"comment":"The stress-energy tensor in Eq. (32) has an incorrect index structure: the second term should be -(1/2)g_{αβ}(∇^μφ∇_μφ+V(φ)), not g_{αβ}(∇^μ∇_μ+V(φ)).","section":"§11, Eq. (32)"}],"recommendation":"major_revision","confidential_remarks":"The paper is readable and addresses a worthwhile question, but the central claim is currently overstated. The most serious issue is the treatment of the even n>2, α>0 branch, where f''(0)=0 and the scalar potential is non-analytic at the minimum; this must be resolved before the correspondence can be claimed. I do not see evidence of misconduct; the problems are technical and appear fixable in a major revision. The authors should either prove nonlinear stability for the cuspy minima or restrict the claimed correspondence to models with f''>0 everywhere, and they should correct the sign errors in Eqs. (19) and (22)."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThis paper works through the conformal transformation from f(R) gravity to scalar-tensor theory and extracts stability conditions for the polynomial families R+αR^n and R+αR^n+βR^m. The algebra for the potential is mostly correct, and the conclusion that stable models require even powers and positive coefficients matches the standard f'(R)>0, f''(R)>0 criterion for the examples shown. The n=2 Starobinsky case is handled correctly. That's about the extent of the good news.\n\nThe central claim—'one-to-one correspondence of scalar sector instability with Ostrogradsky instability'—is not established. The stability criterion the authors use is that V(σ) is real, bounded below, and has at least one minimum. That is too weak. For even n>2 with α>0, the minimum at σ=0 is a cusp: V(σ) ~ |σ|^{n/(n-1)}, so V'' diverges at the minimum and the linearized perturbations are undefined. This happens precisely because f''(0)=0, so the map R↦σ is not a local diffeomorphism there. The paper never checks the Hessian or the validity of the field redefinition, so the classification of R+αR^4 (and higher even powers) as stable is not supported.\n\nThere's a second, related issue: the conformal transformation requires f'(R)>0 globally, but for even n>2 and positive α, f'(R)=1+nαR^{n-1} goes negative for large negative R. The authors work as if the mapping covers all of σ, but it only covers the region where f'>0. That's a gap they don't acknowledge.\n\nAlso, the generality is overstated. Sections 4–9 treat only polynomial f(R) with integer powers, yet Section 10 concludes a correspondence for 'higher curvature theories' in general. And Eq. (19) contains a sign error (should be 2f−A f' over (f')^3, not 2f+A f'), which happens not to affect the extremum condition but is a typo that should be fixed.\n\nIf this were just a pedagogical note for the Starobinsky model, it would be fine. As a general stability theorem, it fails. The useful part is the explicit demonstration for R+αR^2, which is textbook. I don't see a new result here beyond what's in the cited reviews.\n\nMy recommendation: don't publish as is. Either send it back for major revision addressing the cusp and the domain of f', or desk-reject as a routine exercise. It's not an important paper, but a good referee would catch the cusp issue, so I wouldn't object to sending it out if the editor wants to be careful. I'd cite it only for the worked example, not for the correspondence.","headline":"A clean re-derivation of known f(R) stability conditions for polynomial models, but the claimed one-to-one correspondence is undercut by the cuspy minimum for even n>2 and by overgeneralization beyond the cases actually checked.","tokens_in":8590,"tokens_out":7369,"would_cite":false,"duration_ms":71139,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper argues that Ostrogradsky instability in f(R) gravity appears exactly as instability of the dual scalar-tensor potential: for polynomial f(R), stability holds only for even powers with positive coefficients.","keywords":["f(R) gravity","Ostrogradsky instability","scalar-tensor duality","conformal transformation","Einstein frame","Starobinsky model","scalar potential stability","higher-curvature gravity"],"falsifier":"Take $f(R)=R+\\alpha R^4$ with $\\alpha>0$. Near $\\sigma=0$ the potential behaves as $V(\\sigma)\\sim|\\sigma|^{4/3}$, so $V''(0)$ diverges; solving the perturbed scalar equation around $\\sigma=0$, or computing the fluctuation determinant, would show whether small fluctuations have well-defined frequencies. If those fluctuations are ill-defined or grow, the classification of this model as stable would be overturned.","tokens_in":7529,"feed_emoji":"🌌","tokens_out":4801,"duration_ms":44913,"temperature":0.7,"pith_summary":"Higher-curvature f(R) gravity can suffer Ostrogradsky instability from higher derivatives, while its conformally dual scalar-tensor theory contains a scalar field with potential $V(\\sigma)$. This paper tries to prove that the two notions of instability coincide: the scalar sector is stable exactly when the f(R) theory avoids Ostrogradsky instability. For $f(R)=R+\\alpha R^n$, the dual potential is real, bounded below, and has a minimum only for even $n$ and $\\alpha>0$; for $f(R)=R+\\alpha R^n+\\beta R^m$, stability requires even $m,n$ with $\\alpha,\\beta>0$. These are precisely the parameter choices with $f'(R)>0$ and $f''(R)>0$ for all $R$, the standard Ostrogradsky-stability conditions. If correct, stability of a polynomial f(R) model can be read off directly from the shape of the Einstein-frame scalar potential.","feed_headline":"Stability of f(R) gravity mirrors its scalar-tensor dual","feed_subtitle":"For polynomial models, only even powers with positive coefficients pass both stability tests.","key_machinery":"The central object is the Einstein-frame scalar potential $V(\\sigma)$ obtained from the auxiliary-field action by a conformal transformation, with the scalar defined through $f'(R)=e^{(2-d)\\sigma/2}$. The two load-bearing conditions are $f'(R)>0$, which keeps the scalar kinetic term in the correct sign, and positivity of $f''(R)$, which is equivalent to the even-power/positive-coefficient conditions for the polynomial models. The extrema of $V$ are located by $A f'(A)=2f(A)$, and the minimum condition involves the sign of $f''(A)$ at the extremum.","core_discovery":"Starting from the auxiliary-field form of the f(R) action and the conformal transformation $g_{\\mu\\nu}\\to e^{\\sigma}g_{\\mu\\nu}$, the paper derives the Einstein-frame potential $$V(\\$\\sigma$)=\\frac{f'(A)A-f(A)}{(f'(A))^{d/(d-2)}},$$ with $f'(A)=e^{-\\sigma}$ in four dimensions, and takes stability of the dual theory to mean $V$ is real, bounded below, and has at least one minimum. For $f(R)=R+\\alpha R^n$, the derived potential is imaginary for odd $n$, and for even $n$ it has a minimum at $\\sigma=0$ only when $\\alpha>0$, whereas negative $\\alpha$ makes it unbounded below. The two-term extension $R+\\alpha R^n+\\beta R^m$ is stable in the scalar sector only when both exponents are even and both coefficients are positive. Since these are exactly the cases in which $f''(R)>0$ everywhere, with $f'(R)>0$ ensured by the conformal choice, the paper concludes a one-to-one correspondence between scalar-sector instability and Ostrogradsky instability.","pith_inferences":["Applied beyond polynomials, the paper's criterion would predict that any f(R) whose Taylor expansion contains only even powers beyond $R$ with positive coefficients is stable; this is a testable extension but not proven here.","For even $n>2$ and $\\alpha>0$, the minimum of $V(\\sigma)$ at $\\sigma=0$ is a cusp with $V\\propto|\\sigma|^{4/3}$, so cosmological perturbation theory around that vacuum may require a non-standard treatment that the paper does not provide.","The demonstrated one-to-one correspondence holds for polynomial f(R) with integer powers; for non-integer powers or functions where $f'(R)$ vanishes somewhere, the conformal mapping breaks down and the correspondence may fail."],"forward_implications":["For polynomial f(R) models, checking the Einstein-frame scalar potential is equivalent to checking $f'(R)>0$ and $f''(R)>0$.","The Starobinsky model $R+\\alpha R^2$ is stable for positive $\\alpha$ and unstable for negative $\\alpha$, matching the scalar-potential analysis.","Models with odd powers of $R$ or negative coefficients are excluded as stable by both the scalar-sector criterion and the $f''(R)>0$ condition.","For two-term models $R+\\alpha R^n+\\beta R^m$, stability requires both exponents even and both coefficients positive.","The correspondence offers a shortcut: stability of a polynomial f(R) gravity can be assessed from the shape of the dual scalar potential without analyzing higher-derivative equations directly."],"supporting_citations":[{"why":"Supplies the Starobinsky R+R^2 inflation model that serves as the n=2 stable benchmark.","marker":"[8]"},{"why":"Defines the Ostrogradsky theorem that identifies the higher-derivative instability the paper maps to scalar-sector instability.","marker":"[10]"},{"why":"Provides the standard construction of the dual scalar-tensor theory from a higher-curvature action.","marker":"[11]"},{"why":"Establishes the equivalence between the original f(R) action and the auxiliary-field form used to derive the potential.","marker":"[12]"},{"why":"Gives the conformal transformation formula for the Ricci scalar that produces the Einstein-frame action.","marker":"[14]"},{"why":"Supplies the multi-field Ostrogradsky analysis behind the f'(R)>0 and f''(R)>0 stability conditions.","marker":"[19]"}],"fun_headline_variants":["f(R) stability mirrors scalar-tensor dual exactly","Even powers, positive coefficients: f(R) stability rule","Ostrogradsky instability linked to scalar sector in f(R)","Polynomial f(R) stable only with even n and positive α"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument equates stability of the scalar-tensor dual with three properties of the potential — real-valued, bounded below, and having at least one minimum — and for even $n>2$ the minimum at $\\sigma=0$ is a cusp ($V\\propto|\\sigma|^{4/3}$), so the paper assumes this cuspy minimum can serve as a stable vacuum without analyzing perturbations around it.","fun_headline_variants_meta":{"raw":{"variants":["f(R) stability mirrors scalar-tensor dual exactly","Even powers, positive coefficients: f(R) stability rule","Ostrogradsky instability linked to scalar sector in f(R)","Polynomial f(R) stable only with even n and positive α"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000301,"raw_usage":{"total_tokens":1683,"prompt_tokens":843,"completion_tokens":840,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":459,"completion_tokens_details":{"reasoning_tokens":781}},"tokens_in":459,"tokens_out":840,"duration_ms":7826,"temperature":1.0,"reasoning_tokens":781,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T20:07:02.287871+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take $f(R)=R+\\alpha R^4$ with $\\alpha>0$. Near $\\sigma=0$ the potential behaves as $V(\\sigma)\\sim|\\sigma|^{4/3}$, so $V''(0)$ diverges; solving the perturbed scalar equation around $\\sigma=0$, or computing the fluctuation determinant, would show whether small fluctuations have well-defined frequencies. If those fluctuations are ill-defined or grow, the classification of this model as stable would be overturned.","supporting_citations":[{"cited_title":"The scalar-tensor theory of gravitation","cited_arxiv_id":null,"evidence_quote":"Provides the standard construction of the dual scalar-tensor theory from a higher-curvature action."},{"cited_title":"Modified gravity with negative and positive pow- ers of curvature: Unification of inflation and cosmic acceleration","cited_arxiv_id":null,"evidence_quote":"Establishes the equivalence between the original f(R) action and the auxiliary-field form used to derive the potential."},{"cited_title":"Conformal transformations and conformal invariance in gravitation","cited_arxiv_id":null,"evidence_quote":"Gives the conformal transformation formula for the Ricci scalar that produces the Einstein-frame action."},{"cited_title":"Ostrogradsky in theories with multiple fields.Journal of Cosmology and Astroparticle Physics, 2016(06):041, 2016","cited_arxiv_id":null,"evidence_quote":"Supplies the multi-field Ostrogradsky analysis behind the f'(R)>0 and f''(R)>0 stability conditions."}],"review_version":1}