{"id":"814e2536-3586-4a38-bb71-03ddf1a174ca","arxiv_id":"2501.16441","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"A one-loop Schwinger-Keldysh Wilsonian RG calculation generates a dissipative cross-coupling between time branches and predicts two reduced-space fixed points in d=4, related to the Gaussian and Wilson-Fisher fixed points.","lead":"The paper derives, at one loop, how a hot scalar phi^4 theory on the Schwinger-Keldysh time contour generates a dissipative coupling between its two time branches. It then reports two fixed points in a reduced space of complex couplings in d=4, related to the Gaussian and Wilson-Fisher fixed points.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The claimed d=4 fixed points are controlled by a free regulator parameter kappa in Eq. (63); if the physical thermal width differs from the ansatz, the fixed points likely move or disappear.","rationale":"The central claim is the existence of two dynamical fixed points in the reduced CTP coupling space at d=4. The fixed-point equations (90)-(91) depend crucially on the O(g^0) term proportional to i/kappa in Eq. (91). This term originates entirely from the ad hoc replacement eta -> 2 E_k kappa g^2 in Eq. (63), where kappa is left unspecified. Because the fixed-point values scale as kappa^{-1/2} in Eq. (100), any change in the functional form or value of kappa alters the fixed-point structure. The paper is transparent about this limitation, and the conclusion explicitly calls for a derivation of kappa. I checked the reduced-space projection: at tree level, single-branch correlators indeed have vanishing derivatives with respect to g_x and bar-g, so the reduced Callan-Symanzik equation is consistent at one loop. I also verified the low-temperature algebra leading to Eq. (100), including the mass-coupling relation, and found no internal inconsistency. The remaining load-bearing concern is therefore regulator dependence, not a mathematical error. A concrete computation of the physical thermal width would settle whether the fixed points survive once kappa is fixed by the underlying QFT. Since the reader already assigned CONDITIONAL on exactly this basis, the verdict should remain unchanged.","tokens_in":21828,"tokens_out":13832,"duration_ms":122439,"concrete_test":"Compute the one-loop retarded self-energy for the same scalar phi^4 theory at finite T, extract its imaginary part on shell to obtain the physical thermal width Gamma_k(E_k, T), and replace Eq. (63) by eta -> 2 E_k Gamma_k. Re-derive the beta functions (85)-(87), or the dimensionless Eqs. (90)-(91), with this momentum-dependent regulator and solve d_s m^2 = d_s g = 0 in d=4. If no simultaneous zero exists, or if the zeros move beyond the low-temperature perturbative regime, the claimed d=4 fixed points are regulator artifacts rather than robust predictions.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section III.B introduces the pinching-pole regulator by replacing the infinitesimal eta with 2 E_k kappa g^2 in Eq. (63), with kappa a free phenomenological constant. This replacement is the sole source of the O(g^0) terms in the beta functions (85)-(87). The fixed points in the reduced space are then solutions of d_s m^2 = d_s g = 0 in which this O(g^0) term balances the O(g^2) term; their location scales as kappa^{-1/2} in Eq. (100). The existence and position of the d=4 fixed points therefore depend on a specific functional form and constant that have not been derived from the microscopic theory. A physical thermal width Gamma_k from the imaginary part of the self-energy would generally be momentum- and temperature-dependent, not 2 E_k kappa g^2 with constant kappa, so the regularized loop integrals would not reduce to Eq. (64), and the fixed-point equations could acquire different roots or none. The paper's own conclusion (Sec. V) lists the calculation of kappa as outstanding. The secondary premise, that tree-level single-branch correlators form a closed subsystem despite running g_x and bar-g, is logically consistent at one loop because dG/dg_x = dG/d bar-g = 0 for these correlators, but it does not repair the regulator dependence.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper develops a perturbative Wilsonian renormalisation-group treatment of scalar φ^4 theory at finite temperature in the Schwinger-Keldysh closed-time-path formalism. By integrating out spatial-momentum shells, the authors derive one-loop running equations for the mass and for the quartic couplings on both branches of the contour, including the generation of a complex cross-branch coupling g×. They verify that the resulting effective action satisfies the standard CTP unitarity constraints and then restrict their attention to a 'reduced space' of couplings that govern tree-level single-branch time-ordered correlators. In this projected space they find complex fixed points at d=4, one of which they connect, via the d=4−ε limit, to the Wilson-Fisher fixed point and the other to the Gaussian fixed point.","tokens_in":22133,"tokens_out":11418,"duration_ms":100975,"significance":"If the central fixed-point claim held, the paper would introduce a new notion of criticality for real-time thermal field theory in exactly d=4, and it would provide a microscopic example of the dissipative CTP effective actions proposed in the literature. The derivation of the beta functions and the explicit verification of the unitarity constraints are useful and carefully presented. The paper is also commendably transparent about its own limitations: it states that the full theory has no critical points, that the reduced space is a projection, and that the thermal-width parameter κ remains to be computed from the underlying QFT. However, these limitations are precisely what make the central claim conditional. The fixed points exist only in the projected subsystem and only for a specific ad hoc regularisation of pinching-pole singularities, with positions scaling as κ^{-1/2}. The significance is therefore real but not yet established at the level claimed in the abstract.","major_comments":[{"comment":"The existence and locations of the d=4 fixed points are controlled by the ad hoc replacement η → 2 E_k κ g^2 in Eq. (63). This replacement is the sole source of the O(g^0) terms in the beta functions (85)–(87), and the fixed-point values in Eq. (100) scale as κ^{-1/2}. Because κ is left as a free phenomenological constant, with its calculation listed as outstanding in Section V, the claimed fixed points are not shown to be properties of the underlying ϕ^4 theory: a physical thermal width Γ_k obtained from the imaginary part of the self-energy would generally be momentum- and temperature-dependent, so the pinching-pole integral would not reduce to Eq. (64) and the roots of ∂_s m^2 = ∂_s g = 0 could shift or disappear. I ask the authors to either compute κ from the microscopic theory or demonstrate that the fixed-point structure is stable under physically reasonable variations of the width profile. In addition, because the O(g^0) term is independent of g, the beta function ∂_s g is discontinuous at g=0; the Gaussian fixed point is obtained only by the separate argument that all loops vanish there. This discontinuity should be acknowledged and discussed explicitly, since it is part of the same regularisation issue.","section":"Section III.B, Eq. (63), and Section IV, Eqs. (85)–(91), (100)"},{"comment":"The 'novel fixed points' are not fixed points of the full RG flow: by construction they satisfy only ∂_s m^2 = ∂_s g = 0, while ∂_s \\bar g and ∂_s g× remain nonzero at these points. The Callan-Symanzik equation (88) closes on the selected tree-level single-branch correlators only because those correlators have zero derivatives with respect to \\bar g and g×. This is a legitimate mathematical projection, but the physical interpretation needs stronger support. The paper gestures at a fine-tuned external source that would decouple the two CTP branches, but no such construction is supplied, and the abstract and the summary of Section IV say that the paper demonstrates 'the existence of two novel interacting fixed points' without consistently emphasising that these are fixed points of a projected tree-level flow. I recommend either providing a concrete criterion under which the projection is physically realised, or systematically rephrasing the claims as fixed points of the reduced, tree-level subsystem.","section":"Section IV, Eq. (88) and the definition of the reduced space"}],"minor_comments":[{"comment":"Please verify the prefactor in the analytic fixed-point values: solving Eqs. (98) and (99) at d=4 with α_4 = 1/(2π^2) gives \\tilde m^2_* = 3(1-i)e^{-\\tilde β/2}/(2π^2 \\tilde κ^{1/2}) rather than the stated denominator 4π^2. This may be a typographical factor of 2, but it should be corrected.","section":"Eq. (100)"},{"comment":"The linear (canonical) term in the beta function for \\bar g should be (4-d)\\bar g, not (4-d)g, given the definition \\bar g(s) = e^{(4-d)s}(g_0 + δ\\bar g(s)) in Eq. (82). Please correct this in the full set of beta functions.","section":"Eq. (86)"},{"comment":"The symbol 'α_{4−η}' in Eq. (99) appears to be a typo for α_{4−ε}; as written, the same symbol η is used both for the spectral-width regulator and for the dimension shift. Please use a distinct symbol for the dimension shift throughout.","section":"Eq. (99)"},{"comment":"The sentence 'Their are rather lengthy so We do not state them here explicitly' is grammatically incomplete, and the matrix elements a_ij are not given anywhere. Since the eigenvalues plotted in Figure 3 are a quantitative result, providing the explicit expressions (or at least the defining formulas) would make the paper more reproducible.","section":"Section IV, linearised flow discussion"},{"comment":"The caption contains an incomplete reference: 'the blue dashed curve in Figure should be regarded' should say 'in Figure 2' (or the appropriate figure number).","section":"Figure 4 caption"}],"recommendation":"major_revision","confidential_remarks":"This is an interesting but conditional paper. The construction of the dissipative CTP effective action and the verification of unitarity constraints are solid and publishable contributions. The d=4 fixed-point claim, however, depends on the free parameter κ and on the projection to the reduced space; both are acknowledged by the authors, but the central claim in the abstract is stated more strongly than the evidence supports. I would not recommend acceptance until either κ is computed or the robustness of the fixed points under a physically motivated width profile is demonstrated, or until the claims are systematically reframed as properties of the projected tree-level flow. The paper may be better suited to a journal that values framework-building even when the headline result is conditional."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First thing to know: the paper does a clean, explicit one-loop Wilsonian RG in the Schwinger-Keldysh formalism for finite-temperature phi^4, and it earns its keep by showing how the cross-coupling g× between the two branches is generated from pinching-pole regularization. The derivation of the beta functions, the unitarity constraints (76)-(78), and the relation g×=2i Im g all check out as far as I can see. That part is a useful, concrete contribution to the CTP EFT literature, and it gives microscopic backing to the structure proposed in Ref. [8].\n\nThe soft spot is the one flagged in the stress-test. The d=4 reduced-space fixed points exist because Eq. (63) replaces the infinitesimal eta with 2E_k kappa g^2, and the O(g^0) term in the beta function, which balances the O(g^2) term at the fixed point, comes entirely from that replacement. The fixed-point locations scale as kappa^{-1/2}. Nothing in the paper fixes kappa; the authors list calculating it as outstanding. So calling these 'demonstrated' fixed points is too strong. They are solutions of a specific regulator ansatz. A physical thermal width with different momentum or temperature dependence would move or eliminate them. The reduced-space projection is also a real limitation: g× and bar g keep running, so these are not fixed points of the full effective theory. The authors are transparent about both caveats, which is to their credit, but transparency doesn't make the headline claim robust.\n\nThe low-temperature analytic connection to Gaussian and Wilson-Fisher is neat, and the non-commuting limits of T→0 and epsilon→0 are handled carefully. That material is fine.\n\nWho is this for? People working on non-equilibrium EFTs, CTP/thermal field theory, and dissipative hydrodynamics. The framework and the explicit g× derivation are worth having even if the fixed-point story needs more work. A serious referee should engage with it; the right outcome is probably a revision that either computes kappa from a resummation or substantially softens the fixed-point claims and reframes them as properties of a model regulator. I would send it to peer review, with the kappa dependence flagged as the main issue to resolve.","headline":"Solid one-loop CTP RG derivation of dissipative couplings, but the claimed d=4 fixed points are conditional on a free regulator parameter and a reduced-space projection.","tokens_in":22621,"tokens_out":3687,"would_cite":true,"duration_ms":31971,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["81T17","81T28","82B28"],"pacs":["11.10.Hi","11.10.Wx","05.70.Jk"],"model":"deepseek-v4-flash","headline":"A Wilsonian RG on the closed-time-path contour shows that thermal $\\phi^4$ theory has two interacting fixed points in exactly $d=4$ spacetime dimensions, one a generalised Gaussian and one a generalised Wilson-Fisher fixed point, in the…","keywords":["Wilsonian renormalisation group","thermal field theory","Schwinger-Keldysh formalism","phi^4 theory","pinching-pole singularities","dissipative effective field theory","Wilson-Fisher fixed point","complex fixed points"],"falsifier":"Compute the one-loop thermal self-energy to extract the width $\\Gamma_k$: if $\\Gamma_k$ is not of the form $2E_k\\kappa g^2$ with momentum-independent finite $\\kappa$, the fixed-point values in Eq. (100) will shift or disappear; equivalently, a two-loop single-branch four-point function whose finite part depends on $g_\\times$ through internal lines would falsify the reduced-system closure.","tokens_in":21605,"feed_emoji":"🔥","tokens_out":9349,"duration_ms":77692,"temperature":0.7,"pith_summary":"This paper shows that applying the Wilsonian renormalisation group to the scalar $\\phi^4$ theory at finite temperature, formulated on the Schwinger-Keldysh closed-time-path contour, generates a dissipative effective action in which the two branches of the time contour interact through an influence-functional coupling. The authors then restrict attention to time-ordered correlators built from a single branch and define a reduced space of complex couplings, the mass squared and the quartic coupling. In that reduced space they find two interacting fixed points at exactly $d=4$ spacetime dimensions. One connects, as temperature is lowered, to the Gaussian fixed point, and the other to the epsilon-expansion Wilson-Fisher fixed point of the Euclidean $d=4-\\epsilon$ theory. If the analysis is right, single-time-branch time-ordered correlators of the thermal CTP theory become scale invariant at complex couplings even though the full two-branch theory has no critical point.","feed_headline":"Two new fixed points appear in thermal phi^4 theory","feed_subtitle":"A closed-time-path Wilsonian RG ties them to the Gaussian and Wilson-Fisher fixed points.","key_machinery":"The central machinery is the one-loop Wilsonian effective action on the CTP contour, $$S_{\\rm eff}[\\phi_1,\\phi_2]=\\int d^dx\\left[\\tfrac12(\\partial\\phi_1)^2-\\tfrac12 $m^{2}$\\$phi_1^{2}$-g\\$phi_1^{4}$-\\tfrac12(\\partial\\phi_2)^2+\\tfrac12 $m^{2}$\\$phi_2^{2}$+\\bar g\\$phi_2^{4}$+g_\\times\\$phi_1^{2}$\\$phi_2^{2}$\\right],$$ with couplings running in the scale $\\zeta$. The pinching-pole divergence in the four-point function is regulated by replacing $\\eta\\to 2E_k\\kappa g^2$ (Eq. (63)), which makes $g$ complex and generates $g_\\times=2i\\,\\mathrm{Im}\\,g$; this $\\kappa$-dependent term enters the $\\beta$-functions at $O(g^0)$ and is what produces the nontrivial fixed points. The reduced coupling space is the pair $(\\tilde m^2,\\tilde g)$ appearing in the Callan-Symanzik equation for single-branch correlators, where $\\partial_{\\bar g}$ and $\\partial_{g_\\times}$ act as zero; fixed points are simultaneous zeros of $\\partial_s\\tilde m^2$ and $\\partial_s\\tilde g$, and their linearisation gives complex scaling dimensions $\\Delta_m,\\Delta_g$.","core_discovery":"The paper claims that, although the full one-loop CTP effective field theory for thermal $\\phi^4$ has no interacting fixed points once the RG-generated cross-coupling $g_\\times$ keeps running, a reduced system is critical. The reduction is to time-ordered correlators built from a single branch, say $\\phi_1$, of the closed time path, so that the $\\beta$-functions for $\\bar g$ and $g_\\times$ drop out of the Callan-Symanzik equation. In this two-dimensional complex space of $(\\tilde m^2,\\tilde g)$, the one-loop flow has two simultaneous zeros in exactly $d=4$ spacetime dimensions, located at low temperature at $\\tilde m^2_* = 3(1-i)e^{-\\tilde\\beta/2}/(4\\pi^2\\tilde\\kappa^{1/2})$, $\\tilde g_* = (i-1)e^{-\\tilde\\beta/2}/\\tilde\\kappa^{1/2}$, and at the opposite points $(\\tilde m^2_\\circ,\\tilde g_\\circ) = -(\\tilde m^2_*,\\tilde g_*)$. In $d=4-\\epsilon$, as the temperature is lowered the two branches connect respectively to the Gaussian fixed point and to the Wilson-Fisher fixed point of the Euclidean $\\epsilon$-expansion; this motivates calling the $d=4$ branch a generalised Wilson-Fisher fixed point that exists for single-branch correlators at integer dimension.","pith_inferences":["A testable extension not undertaken here is to compute $\\kappa$ from the resummed one-loop self-energy; if the physical thermal width has a momentum or coupling dependence different from $2E_k\\kappa g^2$, the fixed-point locations will shift and the $d=4$ statement will need revision.","The complex, exponentially small fixed-point couplings suggest these are not equilibrium critical points but rather transient scale-invariant regimes or critical points of an associated open system; the paper gestures at the open-QFT interpretation but does not develop it.","A natural check of the reduced-space closure is a two-loop calculation of a single-branch four-point function: if $g_\\times$ enters through internal loops, the reduced Callan-Symanzik system is not closed and the fixed points would be an artifact of the one-loop truncation."],"forward_implications":["At $d=4$, single-branch time-ordered correlators of the thermal CTP effective theory become scale invariant at the two complex reduced couplings $(\\tilde m^2_*,\\tilde g_*)$ and $(\\tilde m^2_\\circ,\\tilde g_\\circ)$, with the fixed-point values vanishing non-analytically as $e^{-\\Lambda/T}$.","In $d=4-\\epsilon$ at fixed low temperature, the two branches of fixed points separate: $(\\tilde m^2_\\circ,\\tilde g_\\circ)$ approaches the Wilson-Fisher point $(-\\epsilon/3, 2\\pi^2\\epsilon/9)$ and $(\\tilde m^2_*,\\tilde g_*)$ approaches the Gaussian point as $T\\to0$, which justifies calling the $d=4$ objects generalised Gaussian and Wilson-Fisher fixed points.","The one-loop effective action satisfies the unitarity constraints (76)-(78), so the dissipative coupling $g_\\times$ is not arbitrary but fixed by $g_\\times=2i\\,\\mathrm{Im}\\,g$; on shell, $\\phi_1=\\phi_2$ and the imaginary parts cancel, leaving a real classical solution.","The RG-generated cross-coupling $g_\\times$ between the two branches appears already at one loop despite being absent in the bare action, giving a concrete microscopic derivation of influence-functional EFTs used for dissipative hydrodynamics."],"supporting_citations":[{"why":"Defines the dissipative CTP EFT structure and unitarity constraints that the one-loop Wilsonian action reproduces.","marker":"[8]"},{"why":"Provides the CTP finite-temperature formalism, KMS relation, and spectral-function representation used throughout.","marker":"[2]"},{"why":"Identifies the breakdown of perturbative thermal spectral densities that underlies the pinching-pole divergence.","marker":"[36]"},{"why":"Establishes the pinching-pole obstruction in finite-temperature transport calculations and motivates the regularisation.","marker":"[37]"},{"why":"Supplies the regularised delta-function identity used to separate the regular and singular parts of the loop integrals.","marker":"[44]"},{"why":"Gives the perturbative form of the thermal width in the spectral function, justifying the replacement $\\eta\\to 2E_k\\kappa g^2$.","marker":"[45]"}],"fun_headline_variants":["Two new fixed points in thermal phi^4 theory","Thermal phi^4 RG discovers two fixed points","CTP approach uncovers two fixed points in phi^4","Wilsonian RG at finite T yields two fixed points","New fixed points from closed-time-path RG in phi^4"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The result rests on two premises: that the pinching-pole regulator can be replaced by a physical thermal width $\\eta\\to 2E_k\\kappa g^2$ with $\\kappa$ an unspecified constant (the $O(1)$ term in $\\kappa^{-1}$ creates the $d=4$ fixed points), and that single-branch tree-level correlators form a closed subsystem even though the full EFT keeps running in $g_\\times$ and $\\bar g$.","fun_headline_variants_meta":{"raw":{"variants":["Two new fixed points in thermal phi^4 theory","Thermal phi^4 RG discovers two fixed points","CTP approach uncovers two fixed points in phi^4","Wilsonian RG at finite T yields two fixed points","New fixed points from closed-time-path RG in phi^4"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000203,"raw_usage":{"total_tokens":1414,"prompt_tokens":999,"completion_tokens":415,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":615,"completion_tokens_details":{"reasoning_tokens":335}},"tokens_in":615,"tokens_out":415,"duration_ms":4300,"temperature":1.0,"reasoning_tokens":335,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T13:15:14.691220+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the one-loop thermal self-energy to extract the width $\\Gamma_k$: if $\\Gamma_k$ is not of the form $2E_k\\kappa g^2$ with momentum-independent finite $\\kappa$, the fixed-point values in Eq. (100) will shift or disappear; equivalently, a two-loop single-branch four-point function whose finite part depends on $g_\\times$ through internal lines would falsify the reduced-system closure.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the CTP finite-temperature formalism, KMS relation, and spectral-function representation used throughout."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the regularised delta-function identity used to separate the regular and singular parts of the loop integrals."},{"cited_title":"Renormalization in Minkowski space-time","cited_arxiv_id":"1908.11311","evidence_quote":"Gives the perturbative form of the thermal width in the spectral function, justifying the replacement $\\eta\\to 2E_k\\kappa g^2$."}],"review_version":1}