{"id":"b9e5fa67-5ea1-484f-8e45-feb6a94e93ef","arxiv_id":"2509.05507","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Extending a stochastic renormalization-group method for tensor group field theories to non-equilibrium, this paper finds that fluctuation-dissipation-violating couplings become large near finite-scale singularities, implying instability of equilibrium dynamics.","lead":"This paper develops a mathematical method for studying out-of-equilibrium dynamics in toy models of quantum gravity called tensor group field theories, adding random fluctuations and tracking how couplings change with distance scale. It finds that tiny deviations from equilibrium can grow dramatically near singular points in the flow, suggesting the equilibrium description becomes unstable.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Unproven \\dot L^{(\\infty)}≈0 (Eq. 5.30) is load-bearing for the Ward-closed β-functions; if false, the bigamous 'fly away' and the 'IR reality' claim may be artifacts.","rationale":"The reader's weakest_assumption correctly identifies the unproven hypothesis \\dot L^{(\\infty)}≈0 (Eq. 5.30) as load-bearing. This assumption is essential to close the Ward-constrained hierarchy and to produce the explicit β-functions (5.58), (5.60) whose numerical integration yields the finite-scale singularity and the bigamous fly-away. My stress-test confirms this is the most concrete and testable point: the term L^{(\\infty)} is a k-dependent function, and there is no analytical argument, only an assertion, for its derivative to vanish. The paper's own caveats about truncation-based UV integrals and about the method's UV focus further undermine the extrapolation to the IR. Since this is exactly the same concern the reader raised, agreement is 'agree'. The appropriate verdict remains conditional: the paper's methodology is promising but the key hypothesis must be verified or replaced before the central physical claim can be accepted. No change to the reader's verdict is needed.","tokens_in":33166,"tokens_out":5807,"duration_ms":60628,"concrete_test":"Compute \\dot L^{(\\infty)}(k) numerically from its definition (5.24) using the truncated propagators (5.2) and the Litim regulator (3.13), for the equilibrium Ward-constrained flow (Δ=Δ'=λ2=0) over the k-range of Figures 14–15. If |\\dot L^{(\\infty)}| is not negligible compared with |\\dot L^{(0)}| (e.g., >10%) at any point before the singularity, then re-derive the Ward constraints (5.35)/(5.41) keeping the full \\dot L, and rerun the flow to see whether the bigamous fly-away (Δ, Δ', λ2) persists. If it disappears, the central claim fails.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim (Section 5.4) that a discontinuous transition to a non-equilibrium IR regime occurs rests on the closed flow equations (5.58) and (5.60). These are obtained from the Ward identity (5.21) only after adopting the hypothesis \\dot L^{(\\infty)}(k)≈0 (Eq. 5.30). But L^{(\\infty)} (5.24) is a k-dependent quantity built from the running propagators and the regulator; there is no demonstrated argument that its logarithmic derivative vanishes. The accompanying statement that Z_∞-suppressed UV integrals can be dropped is not sufficient, because Z_∞ is constant and the derivative of the finite combination can be non-zero. The paper itself warns that truncation-based evaluation of unbounded UV integrals leads to paradoxes, so this term is uncontrolled. If \\dot L^{(\\infty)} is not negligible, the Ward constraints (5.35), (5.41) and the closed expressions for λ' and λ'_2 change, and the numerical singularity/fly-away seen in Figures 14–15 could be an artifact. Additionally, the flow is integrated into the deep IR, beyond the deep-UV regime where the non-branching melonic truncation is justified; the conclusion admits the method's 'focus on the UV regime.' Thus the most load-bearing, testable weakness is the unproven hypothesis (5.30).","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper continues the authors' stochastic-quantization functional renormalization group (FRG) program for group field theories. It studies a rank-5 Abelian U(1) tensorial group field theory with Langevin dynamics, and extends the previous equilibrium analysis to out-of-equilibrium regimes by allowing 'bigamous' interactions that violate the fluctuation-dissipation theorem. The hierarchy of melonic flow equations is closed using Ward identities derived in the symmetric phase, yielding explicit beta functions for the mass, quartic couplings, anomalous dimension, and the new bigamous couplings (Eqs. (5.3), (5.5), (5.15), (5.16), (5.58), (5.60), with the Ward constraints (5.35) and (5.41)). Numerical integration shows that a small FDT-breaking perturbation remains small until the equilibrium flow hits a finite-scale singularity, at which point the bigamous couplings 'fly away' and then partly relax, leaving one coupling (\\(\\bar\\Delta\\)) at a constant nonzero value. The paper interprets this as evidence for a discontinuous phase transition to a non-equilibrium IR regime and states (Claim 1) that no reliable global fixed point exists in the bigamous non-branching melonic sector at leading order of the derivative expansion.","tokens_in":33503,"tokens_out":7340,"duration_ms":74714,"significance":"The paper is a methodological contribution: it constructs a Ward-constrained closure of the melonic FRG hierarchy in a nonequilibrium stochastic setting, with explicit equations and a clearly stated toy model. The equilibrium limit is correctly recovered, and the authors are candid about several limitations of the derivative expansion and the non-branching sector. If the key hypothesis used to close the Ward identities can be justified, the method could become a useful toolbox for nonequilibrium TGFTs. However, the central physical conclusion—that a discontinuous transition to a non-equilibrium IR regime occurs—rests on an unproven hypothesis about the scale derivative of a UV Ward-identity contribution, and on numerical ODE solutions whose method and tolerances are not given. These issues are load-bearing and need to be addressed before the main claim can be accepted.","major_comments":[{"comment":"The hypothesis \\(\\dot L^{(\\infty)}(k)\\approx 0\\) is load-bearing. The Ward identity (5.21) is differentiated to obtain (5.28), and the subsequent constraints (5.35), (5.41) and the closed expressions for \\(\\bar\\lambda'\\), \\(\\bar\\lambda'_2\\) in (5.58) and (5.60) all rely on dropping \\(\\dot L^{(\\infty)}\\) from \\(\\dot L=\\dot L^{(\\infty)}+\\dot L^{(0)}\\). But \\(L^{(\\infty)}\\) in (5.24) is a k-dependent integral containing \\(Z_\\infty\\), the running propagators and \\(\\lambda(k)\\); the derivative acts on these k-dependent objects, so 'Z_\\infty-suppressed UV integrals' does not imply that the logarithmic derivative vanishes. The paper itself warns (Section 4, after Eq. (4.13)) that evaluating unbounded UV integrals with the truncation leads to paradoxes. If \\(\\dot L^{(\\infty)}\\) is not negligible, the Ward constraints change and the numerical fly-away of \\(\\bar\\lambda_2,\\bar\\Delta,\\bar\\Delta'\\),","section":"§5.4, Figures 14–16; §3.3, Eq. (3.15)"},{"comment":"The central numerical conclusion is obtained by integrating the flow across and beyond the finite-scale singularity, into a regime where the non-branching melonic truncation and the derivative expansion are not justified. Section 3.3 states that the non-branching melonic sector is stable in the deep UV regime, and the Conclusion concedes the method's 'focus on the UV regime.' Yet Figures 14–15 show the flow through \\(-\\ln k\\approx 0.35\\) and beyond, and the 'IR reality' claim refers to the large-scale regime. No control parameter (for example, comparison with subleading bubbles or next order in the derivative expansion) is provided in the singularity region. The fly-away could be a truncation artifact. The authors should specify the domain of validity of the flow equations and check whether the fly-away persists under controlled extensions of the truncation.","section":"§5.4, numerical method"},{"comment":"The decisive phenomenon—the fly-away of \\(\\bar\\lambda_2,\\bar\\Delta,\\bar\\Delta'\\) at the equilibrium singularity and the subsequent relaxation—is demonstrated only by numerical ODE integration. The paper gives no integration method, step-size control, tolerances, or code; the initial perturbation is fixed at \\(10^{-8}\\) without discussion of sensitivity. Near a singularity, numerical blow-up and genuine divergence are hard to distinguish, and the conclusion that the system 'avoids' the singularity depends on the trajectory passing through a stiff region. Please provide numerical details and a sensitivity analysis (varying the initial \\(\\bar\\lambda_2\\) from \\(10^{-10}\\) to \\(10^{-6}\\) and varying solver tolerances), or make the code available.","section":"§5.4, final paragraph"},{"comment":"The inference from the observed flow to a discontinuous phase transition is underevidenced. The text itself states that the interpretation is 'indirect' and 'based on an analogy.' A discontinuous transition normally requires an order parameter with a discontinuity across a control parameter; the presented trajectory merely shows that a tiny FDT-breaking coupling changes the flow near a singularity, leaving one coupling constant at a small value. The statement 'likely corresponds to the IR reality of the system' goes beyond what a truncated, sector-restricted toy model can support. Please either weaken the conclusion to 'instability of the equilibrium truncation' or provide additional diagnostics (e.g., a phase diagram in the initial-condition plane or an order-parameter analysis).","section":"§5.4, final paragraph"}],"minor_comments":[{"comment":"The first term on the right-hand side should likely be \\(2\\bar\\Delta'\\) rather than \\(2\\bar\\Delta\\), for consistency with Eq. (5.8).","section":"Eq. (5.17)"},{"comment":"The sentence 'In (5.4), only the two first ones diagrams ...' refers to an equation in Section 3.5; it should refer to Eq. (3.26) (or (3.33)), not to Eq. (5.4).","section":"Section 3.5, after Eq. (3.33)"},{"comment":"'η is again given by equation (5.76)' should refer to Eq. (5.10) (or (5.9)), not to a later equation in Section 5.5.","section":"Section 5.2, after Eq. (5.3)"},{"comment":"'Figures 14, 14, and 14 summarize the main results' should read 'Figures 14, 15, and 16.'","section":"Section 5.4, first paragraph"},{"comment":"The deep-UV condition is printed as \\(\\Lambda\\ll k\\ll 1\\), which is dimensionally inconsistent with \\(\\Lambda\\) a UV cutoff. Presumably \\(\\Lambda\\gg k\\gg 1\\) (in appropriate units) is intended; please correct.","section":"Eq. (3.15)"},{"comment":"The citation '[samary2014closed]' is not a numbered reference in the bibliography; it should be replaced by the proper numbered entry.","section":"End of Section 4, first paragraph"}],"recommendation":"major_revision","confidential_remarks":"The paper is heavily self-referential, importing results from [16,27,28] and inheriting their approximations. The stress-test concern about Eq. (5.30) is genuine and directly affects the central conclusion. The numerical part needs to be reproducible. The fit to the journal is acceptable for a methodological paper, but the main interpretive claim should be made conditional on the unproven hypothesis."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This is a genuine continuation of the authors' FRG program for TGFTs, moving from equilibrium to non-equilibrium by adding FDT-breaking couplings (Δ, Δ′, λ2) and closing the flow hierarchy with Ward identities. The new beta functions for these 'bigamous' couplings are derived in detail, and the observation that they 'fly away' near finite-scale singularities is a plausible and interesting phenomenon. The paper is also honest: it flags its own approximations, notes a typo in a previous reference, and explicitly says the method focuses on the UV regime.\n\nThe main soft spot is exactly where the stress-test note points. The closed flow equations for η, λ′, and λ2′ depend on the hypothesis \\dot L^(∞) ≈ 0 (Eq. 5.30). That hypothesis is stated as 'reasonable' but not derived, and it is load-bearing: it is what converts the Ward identity (5.21) into the explicit beta functions that produce the numerical fly-away in Figures 14–15. The fact that Z∞ suppresses the integral does not by itself make its logarithmic derivative vanish; L^(∞) is a k-dependent object built from running propagators and the regulator. If \\dot L^(∞) is not negligible, the closed flow changes and the claimed discontinuous transition could be an artifact. To the authors' credit, they state the assumption openly rather than burying it, but it remains unsupported.\n\nThe second soft spot is the numerical integration. The paper's main evidence is the behavior of ODE flows, yet no code, method, or tolerances are provided. For a methodological paper, reproducibility of the numerics is not a luxury; it is part of the method.\n\nA third concern, acknowledged in the conclusion, is that the flow is integrated into the deep IR, beyond the regime where the non-branching melonic truncation is justified. The central physical claim—that the IR reality is a non-equilibrium regime—therefore rests on an uncontrolled extrapolation. This does not make the claim false, but it makes it a conjecture supported by an approximation whose validity is not established at the relevant scales.\n\nWho should read this? Specialists in TGFT and FRG, particularly those working on stochastic quantization of tensor models. The paper is not for a general audience and does not settle the physics, but it is a serious technical contribution with clearly identified machinery. I would send it to peer review, expecting major revision: the authors should either prove or convincingly justify \\dot L^(∞) ≈ 0, and provide reproducible numerical details. I would not cite it in my own work unless I were actively working in this niche.","headline":"A credible methodological extension of the authors' equilibrium FRG for TGFTs to non-equilibrium dynamics, with new Ward-closed beta functions for FDT-breaking couplings; but the central claim of an IR non-equilibrium transition rests on an unproven derivative hypothesis and thin numerics.","tokens_in":33983,"tokens_out":1926,"would_cite":false,"duration_ms":21989,"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":"This paper argues that a stochastically quantized group field theory can leave equilibrium at infrared scales, with fluctuation-dissipation-breaking operators becoming relevant near finite-scale singularities of the renormalization group fl","keywords":["stochastic quantization","tensorial group field theory","functional renormalization group","fluctuation-dissipation theorem","Ward identities","melonic sector","nonequilibrium phase transition","Martin-Siggia-Rose formalism"],"falsifier":"Evaluate the derivative of the ultraviolet Ward-identity kernel directly, using the full truncated propagators instead of setting it to zero, and rerun the numerical flow near the singular scale; the claimed transition is settled by whether the bigamous couplings still fly away. An independent check is to include a single branched-melonic contribution and see whether the finite-scale singularity still resolves.","tokens_in":33036,"feed_emoji":"⚛️","tokens_out":7788,"duration_ms":80777,"temperature":0.7,"pith_summary":"This paper develops a renormalization-group formalism for a stochastically quantized tensorial group field theory (a non-local field theory motivated by quantum gravity) without assuming the dynamics stays in equilibrium. The authors work with a just-renormalizable Abelian toy model and extend their earlier equilibrium setup by adding 'bigamous' operators, interactions with two response fields, which violate the fluctuation-dissipation theorem. Using the dominance of non-branching melonic diagrams together with Ward identities, they close the infinite flow hierarchy and obtain closed beta functions for the bigamous couplings. Their numerical solutions show that, along trajectories whose equilibrium flow would hit a finite-scale singularity, an infinitesimal fluctuation-dissipation violation stays small until the singular scale and then grows strongly, temporarily resolving the singularity; the paper interprets this as a discontinuous phase transition to a non-equilibrium regime that may be the model's infrared reality. The paper also claims that no reliable global fixed point exists in the bigamous non-branching melonic sector at leading order of the derivative expansion.","feed_headline":"Group field theory's RG flow can end in a non-equilibrium phase","feed_subtitle":"Small violations of equilibrium relations grow at finite-scale singularities, hinting the infrared reality is out of equilibrium.","key_machinery":"The load-bearing machinery is the non-branching melonic sector together with the Ward identities of the stochastic theory. Melonic diagrams dominate the ultraviolet flow, and the effective vertex expansion expresses the sextic couplings and momentum derivatives through the quartic couplings using Schwinger-Dyson relations; Ward identities constrain the same quantities, closing the hierarchy. The new element is the 'bigamous' parametrization, interactions with two response fields, with couplings that break the fluctuation-dissipation theorem. The computation of the anomalous dimension and of the bigamous beta functions uses a Litim-type regulator while keeping the regulator itself time-revers","core_discovery":"The central claim is that the infrared physics of the toy tensorial group field theory is generically not equilibrium physics. In the symmetric phase, non-branching melonic diagrams dominate, and Ward identities are available because the kinetic term breaks unitary invariance; together these close the otherwise infinite system of flow equations. Relaxing time-reversal symmetry introduces three new couplings, the bigamous ones associated with two response fields. Solving the closed Ward-constrained flow shows that no reliable global fixed point exists in this sector at leading derivative expansion. More strikingly, along a flow that would end in a finite-scale singularity under equilibrium dy","pith_inferences":["If the fly-away persists beyond the approximation that the ultraviolet part of the Ward-identity kernel is stationary, then equilibrium-based phase diagrams for group field theory condensate cosmology may need revision toward a non-equilibrium infrared phase.","The bigamous coupling that stabilizes at a nonzero constant value could serve as an order parameter for the non-equilibrium phase; studying how its late-time value scales with the initial perturbation would calibrate the transition.","A direct numerical evaluation of the neglected term in the Ward-identity derivative could decide whether the claimed discontinuous transition survives without the simplifying hypothesis.","The analogy with spin-glass and disordered dynamics models suggests a 4PI effective-action formalism might confirm the transition directly, a step the paper leaves for later work."],"forward_implications":["Equilibrium truncations of stochastic group field theories cannot be trusted in regimes where the equilibrium flow is singular; the singularity signals non-normalizability of the equilibrium state rather than a mere approximation failure.","Operators forbidden by perturbation theory can become relevant near the singularity, so truncations that exclude them miss the infrared phase transition.","The absence of a reliable global fixed point at leading derivative expansion implies the infrared is not described by scale-invariant equilibrium criticality in this sector.","The Ward-identity plus melonic closure method is transferable to other group field theories and to stochastically quantized non-local models.","Small fluctuation-dissipation violations do not grow along regular trajectories, so equilibrium dynamics remains stable away from singularities."],"supporting_citations":[{"why":"Predecessor paper supplying the stochastic formalism, equilibrium closure scheme, and Ward-identity setup that this paper extends.","marker":"[27]"},{"why":"Introduces the effective vertex expansion and the use of Ward identities to close the melonic flow hierarchy.","marker":"[28]"},{"why":"Provides the equilibrium Ward-constrained melonic beta functions, the sextic-coupling expressions, and the absence-of-fixed-point analysis that this paper builds on and corrects.","marker":"[16]"},{"why":"Source of the bigamous two-response-field parametrization and the strategy for going beyond fluctuation-dissipation equilibrium.","marker":"[54]"},{"why":"Establishes the exact fluctuation-dissipation theorem and the non-propagation of the response field used throughout the stochastic analysis.","marker":"[41]"},{"why":"Shows the Abelian U(1) rank-five model is just-renormalizable, fixing the toy model and its power counting.","marker":"[40]"},{"why":"Supplies the time-reversal-preserving constraint on frequency regulators used in the truncation.","marker":"[47]"},{"why":"The Litim regulator used to evaluate the loop integrals and obtain explicit beta functions.","marker":"[48]"}],"fun_headline_variants":["Group field theory's IR is generically out of equilibrium","Stochastic RG finds no global fixed point for GFT","Toy GFT reveals finite-scale singularities out of equilibrium","Ward identities close GFT flow, expose non-equilibrium","Relaxing time-reversal in GFT leads to singularities"],"cache_read_input_tokens":2688,"weakest_assumption_plain":"The closed flow equations depend on the assumption that the ultraviolet part of the Ward-identity kernel does not change along the flow (Eq. 5.30), and on the non-branching melonic sector remaining the whole theory down to the infrared singularity; if either fails, the fly-away of the bigamous couplings may be an artifact.","fun_headline_variants_meta":{"raw":{"variants":["Group field theory's IR is generically out of equilibrium","Stochastic RG finds no global fixed point for GFT","Toy GFT reveals finite-scale singularities out of equilibrium","Ward identities close GFT flow, expose non-equilibrium","Relaxing time-reversal in GFT leads to singularities"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000259,"raw_usage":{"total_tokens":1368,"prompt_tokens":634,"completion_tokens":734,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":378,"completion_tokens_details":{"reasoning_tokens":650}},"tokens_in":378,"tokens_out":734,"duration_ms":7892,"temperature":1.0,"reasoning_tokens":650,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-05T05:24:13.549467+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Evaluate the derivative of the ultraviolet Ward-identity kernel directly, using the full truncated propagators instead of setting it to zero, and rerun the numerical flow near the singular scale; the claimed transition is settled by whether the bigamous couplings still fly away. An independent check is to include a single branched-melonic contribution and see whether the finite-scale singularity still resolves.","supporting_citations":[{"cited_title":"Pedagogical comments about nonperturbative Ward-constrained melonic renormalization group flow","cited_arxiv_id":"2001.00934","evidence_quote":"Provides the equilibrium Ward-constrained melonic beta functions, the sextic-coupling expressions, and the absence-of-fixed-point analysis that this paper builds on and corrects."},{"cited_title":"Functional renormalization group for “p= 2","cited_arxiv_id":null,"evidence_quote":"Source of the bigamous two-response-field parametrization and the strategy for going beyond fluctuation-dissipation equilibrium."},{"cited_title":"Symmetries of generating functionals of Langevin processes with colored multiplicative noise","cited_arxiv_id":"1007.5059","evidence_quote":"Establishes the exact fluctuation-dissipation theorem and the non-propagation of the response field used throughout the stochastic analysis."},{"cited_title":"Frequency regulators for the nonperturbative renormalization group: A general study and the model A as a benchmark","cited_arxiv_id":"1611.07301","evidence_quote":"Supplies the time-reversal-preserving constraint on frequency regulators used in the truncation."}],"review_version":1}