{"id":"35a0ded4-9ce1-4f33-a344-28fce394ded1","arxiv_id":"2411.13574","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"In (3+1)-dimensional projectable Horava gravity, a unique ultraviolet fixed point produces renormalization group trajectories covering the entire unitary range of the kinetic coupling and predicts a large hierarchy between the Planck mass and the Lorentz violation scale.","lead":"This paper maps all fixed points and flow lines of the renormalization group equations in projectable Horava gravity, a simplified quantum gravity theory. It finds one distinguished ultraviolet fixed point whose trajectories reach the general relativity limit and imply that the scale of Lorentz violation is naturally far below the Planck mass.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Hierarchy claim rests on one-loop beta functions in the large-coupling region near fixed point B; no estimate shows two-loop suppression, so the prediction is not yet controlled.","rationale":"The reader's weakest assumption is exactly the quantitative reliability of the one-loop beta functions near fixed point B and in the deep IR. This is the load-bearing element of the paper because the entire hierarchy claim and the uniqueness of the A->B->lambda->1+ flow are derived by integrating these beta functions through a region where dimensionless couplings are large (us~440, v1~-1.36e4). The paper explicitly invokes large couplings to exclude trajectories from F3-F5, but does not apply the same standard to B. Imposing only G<1 does not guarantee small loop corrections when the beta functions contain high powers of us and va; the true expansion parameter could be G times these large factors. A two-loop computation, or at least a reliable estimate of the leading two-loop terms, would settle whether the one-loop trajectory and the exponent kappa_II, and hence the MLV/MPl hierarchy, survive higher-order corrections. Since this check has not been performed, the conditional verdict is appropriate; my analysis does not change it.","tokens_in":28074,"tokens_out":9665,"duration_ms":100542,"concrete_test":"Compute the leading two-loop contributions to the beta-functions of us and v1 in the background-field gauge using the vertices of the action (6)-(8), and evaluate the ratio of two-loop to one-loop terms along the A->B trajectory for G at the maximal value allowed by weak coupling (G=0.25). If this ratio exceeds 10% at the closest approach to B, the one-loop trajectory and the resulting hierarchy (43) are not under quantitative control; if it is uniformly below 10%, the one-loop prediction is validated.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim—trajectories from point A reaching lambda->1+ with a natural MLV<<MPl hierarchy—passes through the neighborhood of the non-asymptotically-free fixed point B, where Table II gives us=440 and v1=-1.36e4. The universal-trajectory structure (Sec. V B) and the exponent kappa_II=3.84 in Eq. (38b) that controls the suppression of G_IR (Eq. (42), Fig. 11) are computed from the one-loop beta-functions of [25]. The paper excludes the F3-F5 fixed points from trajectory analysis precisely because such large couplings put one-loop validity in question (Sec. V A), yet no analogous justification is given for B. The weak-coupling condition G<1 (Sec. VI B) is insufficient: the one-loop beta-functions contain polynomials up to u_s^9 and v_a^3 (Eq. (16)), so the effective expansion parameter is G times large coupling-dependent factors, which may exceed unity at B for G_max=0.25. The deep-IR behavior (Appendix C) also shows v_a diverging as (lambda-1)^(-1), and the paper itself notes strong coupling there (Sec. VI A); the hierarchy prediction therefore depends on an uncontrolled perturbative regime.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the one-loop renormalization group flow of the marginal couplings of projectable Hořava gravity in (3+1) dimensions. Using the beta-functions of their earlier paper [25], the authors numerically locate all real fixed points of the flow: five at finite kinetic coupling λ and eight at λ = ∞ (Tables I and II). They analyze the stability matrices, showing that some have complex eigenvalues and arguing that this need not contradict unitarity. They then construct RG trajectories emanating from the asymptotically free fixed points. They find that all trajectories from the finite-λ fixed points F1, F2 and from the λ = ∞ fixed points 4 and 7 run into strong coupling, while trajectories from the fixed point A pass near the non-asymptotically-free point B and then flow to the GR-like region λ → 1+ or to the boundary λ → 1/3−, covering the full unitary domain. The gravitational coupling G evolves non-monotonically, and requiring G < 1 along the flow leads to a predicted hierarchy M_LV ≪ M_Pl between the Lorentz-violation scale and the effective Planck mass. The central claims are the uniqueness of the fixed point A for phenomenologically interesting IR behavior and the naturalness of this hierarchy.","tokens_in":28368,"tokens_out":6012,"duration_ms":63632,"significance":"If the one-loop input and the numerical classification are reliable, the paper provides the first complete global picture of the RG flow of projectable Hořava gravity in (3+1) dimensions: a distinguished asymptotically free UV completion (point A), a universal trajectory structure, and a falsifiable prediction that the Lorentz-violation scale is naturally many orders of magnitude below the Planck mass. The strengths of the paper include a thorough fixed-point search with multiple cross-checks (removing different equations, scanning in different variables, direct NSolve verification), explicit stability matrices with a thoughtful discussion of complex eigenvalues, and a transparent derivation of the IR asymptotics in Appendix C. The main risk is that the quantitative hierarchy prediction is obtained by integrating one-loop beta-functions through a region of large couplings near the fixed point B, where the validity of the one-loop approximation is not established and is, in fact, explicitly questioned by the authors for other fixed points with similarly large couplings.","major_comments":[{"comment":"The central hierarchy prediction is obtained by integrating the one-loop beta-functions (16) through the neighborhood of the non-asymptotically-free fixed point B, where us ≈ 440 and v1 ≈ -13566 (Table II). The authors exclude the fixed points F3-F5 from trajectory analysis precisely because such large couplings make the one-loop approximation unreliable (§V A), but no analogous justification is given for the trajectories from A that pass near B. The weak-coupling condition G < 1 invoked in §VI B is insufficient because the beta-functions (16) contain powers up to u_s^9 and v_a^3, so the effective loop-expansion parameter is G multiplied by large polynomials of the couplings, which need not be small even when G < 1. Since the exponent κ_II = 3.84 in Eq. (38b) controls the suppression of G_IR in Eq. (42) and Fig. 11, the predicted hierarchy M_LV/M_Pl = sqrt(G_IR) is not yet quantitatively controlled. Please provide an estimate of two-loop corrections in the large-coupling region, or alternatively state explicitly that the hierarchy is a one-loop-level prediction whose reliability near B and in the deep IR is unproven.","section":"§V B, §VI B, Table II, Eq. (38b)"},{"comment":"The paper claims a complete classification of the fixed points, but the evidence is numerical rather than algebraic. The scans are finite (u_t up to 10^8, u_s up to 10^15), and the asymptotic argument in Appendix A only shows that no further solutions exist in the limit at leading order, not as a rigorous proof. The Bezout bound quoted in §III is about 3×10^5 complex roots, so the absence of additional real roots is a numerical inference. Since the uniqueness of the fixed point A for the phenomenologically interesting trajectories is a central claim, the authors should either provide a rigorous certificate (e.g., a Groebner basis computation or interval-arithmetic verification) or rephrase the statements as strong numerical evidence rather than a complete classification.","section":"§III, Appendix A"},{"comment":"The statement that 'apart from the family found in [26], all trajectories quickly run into strong coupling' is not supported for the asymptotically free fixed points F3-F5, whose trajectories are explicitly not analyzed in §V A because of large couplings. While those fixed points lie at λ < 1/3 and therefore cannot reach the λ → 1+ region, the 'all trajectories' claim in the Summary is too strong as written. The paper should either analyze the trajectories emanating from F3-F5, or narrow the claim to the set of fixed points for which the one-loop approximation is considered valid.","section":"§V A, §VII"}],"minor_comments":[{"comment":"There are several typographical errors: 'differring' in the Introduction, 'reparameteraization' in Section II, 'perurbaation theory' in Section VI B, 'asymptoically' in Section VII, and 'he gravitational coupling' in Section VI A.","section":"Throughout"},{"comment":"The critical angle δ is used in the bullet list before its numerical value is defined; please move the definition earlier or add a parenthetical remark.","section":"§V B 4"},{"comment":"The abstract speaks of a 'single universal trajectory', but the paper describes two distinct universal branches, one to λ → 1+ and one to λ → 1/3− (Figs. 4 and 5). Clarify that 'universal' applies separately to each of the two families.","section":"Abstract and §V B 4"},{"comment":"The explicit polynomials P_G_n and P_χ_n are not reproduced in the paper but only referenced to [25]. For a self-contained reading, at least a brief statement of the maximum degrees and the regime of validity of the one-loop computation would be helpful.","section":"§II, Eq. (16)"}],"recommendation":"major_revision","confidential_remarks":"The manuscript relies on the beta-functions of the authors' own earlier paper [25] without independent verification; this is not a defect per se, but it raises the bar for internal consistency checks. The main technical risk is the uncontrolled one-loop approximation in the large-coupling region near point B, which directly affects the headline hierarchy claim. If the authors can reframe that claim as a one-loop prediction with explicit caveats, or supply a perturbative control estimate, the paper could be suitable for publication after revision."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Worth a serious look. This paper completes the one-loop fixed-point census for (3+1)-dimensional projectable Horava gravity: five finite-λ points including new F5, eight points at λ=∞, with stability matrices and a thoughtful discussion of why complex eigenvalues need not contradict unitarity. The numerical search is done honestly: multiple equation-removal strategies, asymptotic arguments near λ=1/3 and us→∞, and a Mathematica NSolve cross-check. The IR asymptotics in Appendix C are genuinely useful, with explicit exponents for us, va, and G as λ→1+. The paper also says plainly where it is not venturing: F3–F5 trajectories are excluded because the one-loop approximation is questionable there, and the deep IR is acknowledged as strong-coupled. That transparency earns credit.\n\nNow the main caveat, which the stress-test note puts exactly right. The trajectories that reach λ→1+ do so by passing near fixed point B, where us≈440 and v1≈−1.36×10^4. The one-loop beta functions contain powers up to us^9 and va^3, so the effective expansion parameter is not simply G but G times large-coupling-dependent factors. The condition G<1 does not obviously control those factors, and the paper gives no two-loop estimate or parametric argument that the neglected terms stay small. The claimed hierarchy MLV≪MPl and the exponent κII=3.84 that controls it are obtained by integrating through precisely that region. So the hierarchy result is conditional on an assumption the paper does not discharge. This is not an internal contradiction; the authors are careful and the claim is a legitimate one-loop prediction. But it is a genuine uncontrolled step, and it should be the referee's first question.\n\nTwo smaller points. First, the central trajectory family from A to λ→1+ already appeared in the authors' 2023 letter [26]; the new value here is the fixed-point classification, the F3–F5 exclusion, the stability analysis, and the G-running/hierarchy discussion. That is enough to justify the paper, but the novelty is incremental rather than a surprise. Second, fixed-point completeness is a numerical claim—strongly supported, but not an algebraic proof. The multiple scanning strategies and the NSolve check make the list credible, yet a skeptical reader cannot independently verify it without code or data, which are not shipped.\n\nWho gets value from this: people working on Horava gravity and on RG flows of Lifshitz-type theories. It deserves a serious referee. The referee should push for an estimate of two-loop contributions or some scaling argument that controls the large-coupling region near B. If the authors can supply that, this becomes the standard reference for the one-loop global flow. Without it, the hierarchy should be treated as an interesting but unproven one-loop prediction.","headline":"A careful one-loop census of the projectable Horava RG landscape that completes the A-to-B-to-λ→1+ picture, but the central hierarchy prediction still rests on one-loop beta functions in a large-coupling regime where control is not demonstrated.","tokens_in":28837,"tokens_out":2712,"would_cite":true,"duration_ms":32962,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["04.60.-m","11.10.Hi"],"model":"deepseek-v4-flash","headline":"Projectable Hořava gravity has exactly one asymptotically free ultraviolet fixed point from which trajectories can reach the general-relativistic regime $\\lambda\\to 1^+$.","keywords":["Horava gravity","projectable Horava gravity","renormalization group flow","asymptotic freedom","fixed points","Lifshitz scaling","Lorentz invariance violation","Planck mass hierarchy"],"falsifier":"Compute the two-loop $\\beta$ functions for the marginal couplings and check whether the fixed points A and B, the connecting stable manifold, and the power-law exponents survive in the large-coupling region near B and in the deep infrared $\\lambda\\to1^+$. If two-loop corrections change the sign of the $\\beta$ function for $G$ near B, move point B, or open a repulsive direction that breaks the trajectory family, the unique-trajectory and hierarchy claims fail. A less expensive check is to repeat the fixed-point search with a different elimination of variables at higher precision, looking specifically for an additional asymptotically free fixed point with $\\lambda>1$.","tokens_in":27886,"feed_emoji":"🌌","tokens_out":10735,"duration_ms":92614,"temperature":0.7,"pith_summary":"This paper asks which renormalization group trajectories a quantum theory of gravity called projectable Hořava gravity can follow, and whether any of them are both consistent in the ultraviolet and look like general relativity at low energies. The authors study the flow of the marginal couplings, the coefficients that stay marginal under the theory's Lifshitz scaling, and claim that among all asymptotically free ultraviolet fixed points only one, located at infinite kinetic coupling $\\lambda$, can serve as a source for trajectories that later reach $\\lambda\\to1^+$, the regime where the kinetic term matches general relativity. Every such trajectory is nearly the same curve in coupling space; only the overall gravitational coupling differs. Along that curve the gravitational coupling first rises, then falls, ending very small in the infrared. Requiring the theory to remain weakly coupled all the way then implies that the Lorentz-violation scale sits far below the Planck mass, a hierarchy that was previously regarded as fine-tuned.","feed_headline":"A single quantum path links Horava gravity to Einstein gravity","feed_subtitle":"The same flow forces the Lorentz-breaking scale far below the Planck mass.","key_machinery":"The machinery is the one-loop $\\beta$-function system for the essential couplings of the theory, in which the gravitational coupling $G$ factorizes, so the flow of $\\{\\lambda, u_s, v_1, v_2, v_3\\}$ can be studied independently. The load-bearing objects are two fixed points at infinite $\\lambda$: point A, the unique asymptotically free fixed point whose repulsive eigenvectors carry the flow toward finite $\\lambda$, and point B, a non-asymptotically-free attractor inside the $\\lambda=\\infty$ plane. The universal trajectory is the stable manifold from A to B followed by ejection along B's single repulsive eigenvector; the ratio of $\\beta$ functions near B yields the steep power law $G \\propto \\lambda^{3.84}$, which converts the condition of weak coupling into a very small infrared $G$ and hence into the Planck-mass hierarchy. The paper also identifies complex eigenvalues of stability matrices at some fixed points and argues these do not contradict unitarity because the deforming operators are not invariant under the linearized gauge symmetry.","core_discovery":"The central discovery is that the renormalization group flow of projectable Hořava gravity in (3+1) dimensions has a unique asymptotically free fixed point, called point A, sitting at infinite kinetic coupling $\\lambda$, whose repulsive directions reach the physically interesting infrared region $\\lambda\\to1^+$. Trajectories leaving point A are first attracted to a second fixed point, point B, which is not asymptotically free but governs intermediate scales; near B the flow scatters and then follows the single repulsive direction of B. After the scattering, all viable trajectories coincide in the space of the couplings $\\{\\lambda, u_s, v_1, v_2, v_3\\}$, and their differences reduce to the initial value of the gravitational coupling $G$. The running of $G$ is non-monotonic: it grows after leaving A, reaches a maximum, and then falls, so that its infrared value $G_{\\rm IR}$ is extremely small whenever the intermediate maximum is held below one for perturbative consistency. The paper concludes that $M_{\\rm LV}/M_{\\rm Pl} = \\sqrt{G_{\\rm IR}} \\ll 1$ arises naturally, rather than by accident.","pith_inferences":["One can infer, beyond the paper, that the same scattering-off-an-intermediate-attractor mechanism may appear in other Lifshitz-type gravity models; if the non-projectable version has an analogous point B, the hierarchy it produces would suppress Lorentz-violating effects in the gravitational sector at low energies.","The one-loop exponent $\\kappa=3.84$ is a natural place to test the claim; if a two-loop computation changes the sign of the beta function near B or shifts the stable manifold, the numerical hierarchy would change, so the quantitative prediction is checkable.","The complex stability eigenvalues may be a general feature of gauge theories at asymptotically free fixed points rather than a peculiarity of Hořava gravity; searching for similar complex eigenvalue pairs in other gauge theories would tell whether the unitarity argument generalizes.","Because the hierarchy is tied to the RG time the flow spends near B, an embedding of this model in a larger theory would need no special parameter choice to explain the smallness of Lorentz violation; the flow itself produces it."],"forward_implications":["If correct, projectable Hořava gravity has a distinguished UV completion: among all asymptotically free fixed points only point A can reach $\\lambda\\to1^+$, so phenomenological modeling can focus on a single family of trajectories.","All trajectories that reach the general-relativistic regime pass through the vicinity of point B and then coincide; low-energy physics of the marginal couplings is essentially universal, with the only free data being the normalization of $G$.","The same flow makes $G$ vanish both in the ultraviolet and the infrared, so the theory is asymptotically free in the UV, while in the IR the vanishing of $G$ does not mean the theory is free because the other couplings grow.","Weak coupling along the flow enforces $G_{\\rm IR}\\ll1$, which translates into $M_{\\rm LV}/M_{\\rm Pl} = \\sqrt{G_{\\rm IR}} \\ll 1$; the Lorentz-violation scale is naturally many orders of magnitude below the Planck mass.","The numerical search finds no asymptotically free fixed points inside the unitary interval $\\lambda>1$ at finite $\\lambda$, and the remaining finite-$\\lambda$ fixed points all lie at $\\lambda<1/3$, so none of them can compete with point A."],"supporting_citations":[{"why":"Supplies the one-loop beta functions for the marginal couplings, the equations whose fixed points and trajectories the paper classifies.","marker":"[25]"},{"why":"Earlier identification of the family of trajectories from the infinite-$\\lambda$ fixed point to the infrared; this paper completes the classification and shows uniqueness.","marker":"[26]"},{"why":"Derives the first-step RG flow equations for projectable Hořava gravity in (3+1) dimensions that underlie the beta functions.","marker":"[24]"},{"why":"Shows (2+1)-dimensional projectable Hořava gravity is asymptotically free, providing the contrast for the absence of finite-$\\lambda$ fixed points with $\\lambda>1$ in 3+1 dimensions.","marker":"[23]"},{"why":"Establishes that the $\\lambda\\to\\infty$ limit is a regular, weakly coupled fixed point, which justifies treating point A as a legitimate UV completion.","marker":"[42]"},{"why":"High-energy scattering amplitudes used to argue that the vanishing of $G$ in the infrared does not restore weak coupling because other couplings grow.","marker":"[43]"},{"why":"Documents strong coupling near $\\lambda\\to1^+$, the reason the flow cannot be allowed to approach the GR-like boundary too closely.","marker":"[33]"},{"why":"Proof of perturbative renormalizability of projectable Hořava gravity, justifying the closed set of marginal couplings used for the RG flow.","marker":"[21]"}],"fun_headline_variants":["One fixed point rules Hořava gravity's flow","Unique quantum path links Hořava gravity to Einstein","Hořava gravity flow reveals a single universal trajectory","A unique fixed point yields natural hierarchy in gravity flow","Gravity flow's unique fixed point sets Lorentz scale"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The calculation assumes that quantum corrections computed to one loop stay reliable even where some couplings grow very large, both near the intermediate fixed point called B and in the deep infrared approach to $\\lambda=1$, because the claimed unique trajectory and its hierarchy are obtained by integrating the flow through exactly those regions.","fun_headline_variants_meta":{"raw":{"variants":["One fixed point rules Hořava gravity's flow","Unique quantum path links Hořava gravity to Einstein","Hořava gravity flow reveals a single universal trajectory","A unique fixed point yields natural hierarchy in gravity flow","Gravity flow's unique fixed point sets Lorentz scale"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000564,"raw_usage":{"total_tokens":2694,"prompt_tokens":982,"completion_tokens":1712,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":598,"completion_tokens_details":{"reasoning_tokens":1636}},"tokens_in":598,"tokens_out":1712,"duration_ms":14168,"temperature":1.0,"reasoning_tokens":1636,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T21:08:54.589946+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the two-loop $\\beta$ functions for the marginal couplings and check whether the fixed points A and B, the connecting stable manifold, and the power-law exponents survive in the large-coupling region near B and in the deep infrared $\\lambda\\to1^+$. If two-loop corrections change the sign of the $\\beta$ function for $G$ near B, move point B, or open a repulsive direction that breaks the trajectory family, the unique-trajectory and hierarchy claims fail. A less expensive check is to repeat the fixed-point search with a different elimination of variables at higher precision, looking specifically for an additional asymptotically free fixed point with $\\lambda>1$.","supporting_citations":[{"cited_title":"Stellar center is dynamical in Horava-Lifshitz gravity","cited_arxiv_id":"0911.1814","evidence_quote":"Establishes that the $\\lambda\\to\\infty$ limit is a regular, weakly coupled fixed point, which justifies treating point A as a legitimate UV completion."},{"cited_title":"Nonlinear superhorizon perturbations in Horava-Lifshitz gravity","cited_arxiv_id":"1105.0246","evidence_quote":"High-energy scattering amplitudes used to argue that the vanishing of $G$ in the infrared does not restore weak coupling because other couplings grow."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Proof of perturbative renormalizability of projectable Hořava gravity, justifying the closed set of marginal couplings used for the RG flow."}],"review_version":1}